Dietary iron intake and cardiovascular outcome in Italian women: 10-year follow-up.

BACKGROUND: Both low and high iron levels have been associated with cardiovascular outcome. The aim of this study was to determine if dietary iron intake is a cardiovascular predictor in women at a population level. METHODS: A 138-item food frequency questionnaire (FFQ) was administered to 906 women aged 61.1±17.1 years from an unselected general population followed for 10 years. The mass of dietary items was calculated, and each food was resolved into its chemical components according to food composition tables conceived for Italian food. RESULTS: An inverse association between iron intake and cardiovascular morbidity was found. Incident nonfatal cerebrovascular events were 10 of 302 (3.3%), 4 of 302 (1.3%), and 2 of 302 (0.7%) in the first, second, and third tertiles of nonheme iron intake (p for trend <0.01), respectively; fatal cerebrovascular events were 20 of 302 (6.6%), 13 of 302 (4.3%), and 5 of 302 (1.7%), respectively (p<0.01); nonfatal coronary events were 24 of 302 (7.9%), 13 of 302 (4.3%), and 10 of 302 (3.3%), respectively, and fatal coronary events were 20 of 302 (6.6%), 15 of 302 (5.0%), and 10 of 302 (3.3%), respectively (p<0.01). Independenly of confounders, hazard ratios (HR) and 95% confidence intervals (CI) of being the first rather than third tertile of nonheme iron intake were 5.00 (1.08-23.26), 3.18 (1.41-10.12), and 3.25 (1.39-7.59), respectively, for cerebrovascular and 2.48 (1.17-5.25), 2.25 (1.03-4.80), and 2.51 (1.39-4.53), respectively, for coronary events. Heme iron intake was not a predictor. CONCLUSIONS: Based on a longitudinal population analysis, low iron intake seems to be an independent predictor of cerebrovascular and coronary morbidity and mortality in women.