Background and aim: Epidemiological studies have suggested an association between higher body iron stores and coronary artery disease (CAD), but recent trials have reported conflicting data on the role of ferritin in CAD. To assess these findings, we examined the association between serum ferritin and the angiographic extent of coronary atherosclerosis in consecutive patients referred for coronary angiography.
Methods and results: We studied 100 consecutive white subjects (41 women and 59 men; mean age 63.7 +/- 11.0 years) who underwent coronary angiography. The data collected at baseline included conventional risk factors for coronary artery disease (CAD), lipid and fasting total homocysteine levels, serum ferritin levels and transferrin saturation, and clinical characteristics. Serum ferritin levels and transferrin saturation (serum iron concentration divided by total iron-binding capacity) were used as measures of the amount of circulating iron available to tissues. Two experienced cardiologists blinded to the clinical and laboratory data reviewed the angiographic cinefilms, and defined the angiographic severity of CAD on the basis of the sum of three vessel scorring systems. The risk of CAD assessed by coronary angiography was not related to ferritin concentrations or transferrin saturation levels. The estimated relative risk of CAD for the fifth vs the first quintile of serum ferritin was 0.83 (95% CI: 0.63-1.24). Forty of the 100 patients had no or minimal CAD (group A; score 0-3), 33 moderate CAD (group B; score 4-8) and 27 severe CAD (group C; score > 8): the serum ferritin levels in the three groups were respectively 165 +/- 126, 167 +/- 121 and 164 +/- 110 ng/ml, and did not represent an independent risk factor for CAD (p = 0.98). Transferrin saturation in the three groups was 22.9 +/- 10%, 21 +/- 9% and 19.9 +/- 10%, with no significant relationship to the severity of CAD (p = 0.23). The presence of angiographic CAD was associated with patient age (p = 0.048), male gender (p < 0.01), high lowdensity lipoprotein cholesterol levels (p = 0.02), low high-density lipoprotein cholesterol levels (p = 0.02), high plasma fibrinogen levels (p < 0.01) and high fasting total homocysteine levels (p = 0.04).
Conclusion: In patients referred for coronary angiography, higher ferritin concentrations and transferrin saturation levels were not associated with an increased extent of coronary atherosclerosis.