NEW YORK (Reuters):

Foods rich in flavonoids - from apples and pears to dark chocolate and red wine - may help shield postmenopausal women from coronary heart disease, cardiovascular disease and stroke, a new study shows.

Flavonoids are antioxidant compounds, found in many plant-basedfoods, and have been hypothesised to protect the heart by reducing levels of low-density lipoprotein or 'bad' cholesterol and reducing inflammation, Dr. Pamela J. Mink of Exponent, Inc., and colleagues note. But studies investigating heart health and flavonoid levels in the diet have had mixed results, they add in the American Journal of Clinical Nutrition.

The researchers used three newly available databases from the United States Department of Agriculture to determine the flavonoid content of foods and analysed results of food questionnaires on diet from 34,489 postmenopausal women participating in the Iowa Women's Health Study.

Mink and colleagues specifically examined the association between the amount of flavonoids the diet and heart disease and death over a 16-year period. The new information allowed the researchers to look at both total flavonoids and seven different subclasses of the plant compound.

Three subclasses of flavonoids, anthocyanidins, flavanones, and flavones, were linked to a significantly reduced risk of heart disease, blood vessel disease or cardiovascular disease mortality, the researchers found. Risk reductions ranged from 10 per cent, for anthocyanidins and heart-related mortality, to 22 per cent, for flavanones and heart disease.

Foods linked to reduction

Specific foods also were linked to risk reductions in heart, blood vessel disease and mortality as well, including bran, which provided a lower risk of cardiovascular disease and stroke; apples, pears, and red wine, which cut both cardiovascular and coronary heart disease risk; grapefruit, which cut coronary heart disease risk, and strawberries and chocolate, tied to a lower risk of cardiovascular disease.

The findings are not definitive, and should be replicated in other prospective studies including large numbers of subjects, the researchers conclude.