Iron and coronary heart disease.

Several strands of evidence suggest a role for iron in heart disease. For example, the massive accumulation of iron that occurs in genetic haemochromatosis and thalassaemia is a well recognised cause of cardiomyopathy. Furthermore, it has been suggested that iron may be associated with coronary heart disease and that this may occur at quite modest concentrations of stored iron.'4 In 1981 Sullivan hypothesised that the difference in the rates of heart disease between men and women could be explained by their different concentrations of stored iron.' Premenopausal women in affluent societies enjoy relative protection from heart disease compared with men, although this relative protection ends with the menopause. Iron stores, as assessed by the serum ferritin concentration, rise steadily in men after adolescence, but in women they remain low before rising after the age of 45.' These changes parallel the incidence of coronary heart disease in the West.2 Additionally, Sullivan postulated that the low rates of heart disease in developing countries may be related to the high prevalence of iron deficiency in these populations.' This hypothesis was supported by Lauffer, who reported a correlation (r = 0 55, p < 0-025) between the median value ofhepatic stored iron and the mortality from ischaemic heart disease in different countries.3 He concluded that it may be possible to identify people at risk of coronary heart disease and reduce this risk by venesection. The low iron stores in some populations may, however, be due to a poor diet: such diets may be less atherogenic because offactors other than iron content. The debate has intensified after the recent report by Salonen et al that modestly raised serum ferritin concentrations increased the risk of acute myocardial infarction.4 In a prospective study of eastern Finnish men, those with a serum ferritin concentration above 200 ,ug/l were more than twice as likely to have an acute myocardial infarction as men with a serum ferritin concentration below this figure. The risk was highest in men who had both a serum ferritin concentration above 200 ,ug/l and a serum low density lipoprotein cholesterol concentration above 5 0 mmol/l. Increased serum ferritin concentrations reflect increased iron stores but may also be related to inflammatory responses and alcohol consumption.5'These variables did not, however, explain the differences in ferritin concentrations or the rates of myocardial infarction.4 Dietary iron intake was also significantly associated with the risk of disease. The proposed mechanism implicating …