Review: Baseline iron levels are not associated with the development of coronary heart disease
ACP J Club. 1999 July-Aug;131:25. doi:10.7326/ACPJC-1999-131-1-025
Danesh J, Appleby P. Coronary heart disease and iron status. Meta-analyses of prospective studies. Circulation. 1999 Feb 24;99:852-4.
Do cohort studies show that iron levels at baseline are associated with an increased risk for coronary heart disease?
Studies were identified by using MEDLINE with search terms related to iron status (iron, ferritin, and transferrin) and coronary heart disease (CHD, myocardial infarction, atherosclerosis, and vascular disease). Bibliographies of relevant studies were scanned, important journals were hand searched, and authors were contacted.
Cohort studies were selected if they had long-term follow-up and reported correlations between iron status at baseline and the development of coronary heart disease.
Data were extracted on size and type of cohort, mean age, follow-up duration, assay methods for iron status (serum ferritin, transferrin saturation, total iron-binding capacity, serum iron levels, or dietary analysis by using questionnaires), other risk factors, adjustment for potential confounders, and coronary heart disease.
12 studies that included 7800 cases of coronary heart disease met the inclusion criteria. Several studies provided data using ≥ 1 test for iron status: 5 studies used serum ferritin, 5 studies used transferrin saturation, 4 studies used total iron-binding capacity, 3 studies used serum iron levels, and 3 studies used dietary questionnaires. Analysis for all studies compared participants in the top third of iron levels at baseline with participants in the bottom third except for studies of serum ferritin, which compared levels of ≥ 200 with < 200 µg/L. All studies but 1 adjusted for smoking and other cardiovascular risk factors. No important association was shown between iron levels measured by any method and coronary heart disease (Table).
Baseline iron levels are not associated with an increased risk for the development of coronary heart disease.
Source of funding: No external funding.
For correspondence: Dr. J. Danesh, Clinical Trials Service Unit, Radcliffe Infirmary, Oxford, England OX2 6HE, UK. FAX 44-1865-558817.
Table. Baseline iron status and coronary heart disease
|Iron status marker||Number of cases||Weighted mean age||Weighted follow-up||Relative risk (95% CI)|
|Serum ferritin||570||55 y||8 y||1.03 (0.83 to 1.29)|
|Transferrin saturation||6194||56 y||14 y||0.92 (0.74 to 1.14)|
|Total iron-binding capacity||2755||58 y||13 y||0.98 (0.66 to 1.46)|
|Serum iron||2848||58 y||14 y||0.83 (0.67 to 1.03)|
|Total dietary intake||2535||59 y||10 y||0.84 (0.66 to 1.06)|
Many epidemiologic studies have examined the relation between coronary heart disease and such biochemical markers as fibrinogen and C-reactive protein levels, antibodies to various infective agents, and iron status. This meta-analysis of prospective studies by Danesh and Appleby concluded that no evidence supports a relation between coronary heart disease and baseline iron levels.
The initial theory that iron levels were associated with coronary heart disease was based on its role in lipid peroxidation. Super-oxide radicals generate iron from ferritin (1), and metal ions seem to be required for the peroxidation of low-density lipoprotein cholesterol by macrophages (2). The studies reviewed in this meta-analysis addressed many types of iron stores, including ferritin levels (which are assumed to be the most accurate); transferrin saturation; iron-binding capacity; and iron intake (through questionnaires). None of the combined analyses found a strong relation between any measure of iron status and the presence of coronary heart disease. This may be because of the effect that inflammation had on iron-carrying proteins and the marked individual variation in levels over time. The possible confounding effect by markers of inflammation is suggested by the known relation to fibrinogen levels, C-reactive protein levels, and leukocyte count reviewed in another meta-analysis done in 1998 by Danesh (an author of this study) and colleagues (3).
Some of the included studies did raise the question of a relation between acute myocardial infarction and iron levels while agreeing that coronary heart disease was not associated with iron levels. The relation between iron and myocardial infarction was not present in men using vitamin E but was higher in those with other traditional risk factors (4). The role of iron in our understanding of coronary heart disease as it relates to oxidative stress, perhaps, has still not been answered.
Robert Weiss, MD
Central Maine Medical CenterAuburn, Maine, USA
3. Danesh J, Collins R, Appleby P, Peto R. Association of fibrinogen, C-reactive protein, albumin, or leukocyte count with coronary heart disease: meta-analyses of prospective studies. JAMA. 1998;279:1477-82.