Flavonoids and heart disease

BMJ 1996; 312 doi: http://dx.doi.org/10.1136/bmj.312.7029.458 (Published 24 February 1996) Cite this as: BMJ 1996;312:458
  1. Matthew F Muldoon,
  2. Stephen B Kritchevsky
  1. Assistant professor of medicine Center for Clinical Pharmacology, University of Pittsburgh School of Medicine, Pittsburgh, PA 15260, USA
  2. Associate professor of preventive medicine Division of Biostatistics and Epidemiology, Department of Preventive Medicine, University of Tennessee, Memphis, TN 38163, USA

    Evidence of benefit still fragmentary

    A longstanding tenet of nutrition holds that people with diets rich in fruits and vegetables enjoy better health than people eating few fruits and vegetables. Consequently, research has sought the components or compounds responsible for this apparent health benefit. Much of that research currently focuses on dietary antioxidants, fuelled by our growing appreciation that damaging oxidative processes are a common biochemical link between otherwise pathophysiologically distinct diseases.1 For example, the development of early atherosclerotic lesions is now thought to be specifically promoted by low density lipoprotein particles that have been oxidatively modified,2 3 and oxidative damage to nucleic acids may have an important role in carcinogenesis.

    Dietary antioxidants, which complement the actions of enzymatic antioxidants, are now widely recognised as including (alpha) tocopherol (vitamin E), ascorbic acid (vitamin C), and β carotene (a precursor of vitamin A). Perhaps flavonoids should now be added to this list. The article by Knekt et al in this issue of the BMJ (p 478)4 joins two other epidemiological reports5 6 in suggesting a role for the flavonoids, and for quercetin in particular, in the prevention of coronary artery disease.

    This Finnish study relates subjects' usual diet, as reported for the year before entry to the study, to mortality from heart disease over the subsequent 26 years.4 The findings show a modestly protective effect, after consideration of other important dietary factors such as saturated fat. (Though not commented on by the authors, flavonoids also seem to be associated with fewer deaths from diseases other than atherosclerosis.) The case for a specific cardioprotective role of quercetin comes from the data showing that apples and onions, important sources of quercetin in the Finnish diet, were the foods most strongly related to mortality risk.4 It is unfortunate that these food-specific analyses were not adjusted for total energy consumption, since an overall increase in calorific intake could easily be a marker of high physical activity or body size, both of which are themselves inversely correlated with risk of heart disease.

    As intriguing as these flavonoid findings are, the epidemiological evidence is not entirely consistent. In the atherosclerosis risk in communities study, consumption of apples was not related to thickness of the carotid artery wall (Kritchevsy, unpublished data), nor was flavonoid intake associated with heart disease in either the nurses health study or the male health professional study (Eric B Rimm, personal communication). It may be that the contribution of a particular food to cardiovascular health depends on the composition of the rest of the diet. The data of Knekt et al show a relatively low intake of vitamin C in the Finnish diet, and, since quercetin may lead to sparing of vitamin C, flavonoids may be most important in populations with marginal vitamin C intake.

    Interpretation of this information by practising clinicians (for themselves or for curious patients) is challenging. One needs to know the biochemical composition of flavonoids, their pharmacology and range of physiological effects, their potential for toxicity, and their established health benefits. Structurally, flavonoids are polyphenols of about the molecular weight of cholesterol; as such, they are small, lipid soluble compounds like (alpha) tocopherol and β carotene. However, unlike these two established dietary antioxidants, flavonoids are ubiquitous in the plant kingdom and include many thousands of individual compounds. Particularly staggering is their list of reported actions.7 In a variety of in vitro assays flavonoids have been found to possess anti-inflammatory, antiallergic, antithrombotic, venotonic, antimicrobial, and antineoplastic activities. Some are oestrogenic, others are antithyroidal, and certain flavonoids can be mutagenic.8 On a molecular level flavonoids can modify the actions of a host of enzymes, including those involved in cell respiration and replication, drug metabolism, and immune function.

    Clinical relevance is unclear

    Some, though not all, of these actions of flavonoids are thought to be due to their antioxidant properties as metal chelators, free radical scavengers, and chain breaking antioxidants.7 Even here though, one should recognise the complexity of interactions between pro-oxidants and antioxidants. A compound may act as an antioxidant in one oxidative challenge system and be neutral or even pro-oxidant in another. For example, whereas quercetin can sequester and reduce the activity of oxidant inducing metals (such as iron and copper), with an anion exchange chromatographic method for testing potential inhibitors of low density lipoprotein oxidation, quercetin was found to have pro-oxidant activity.9

    The clinical relevance of these laboratory findings is unknown. Theoretically, flavonoids may prevent coronary artery disease by inhibiting low density lipoprotein oxidation, by reducing platelet aggregation, or by reducing damage from ischaemia and reperfusion. It should also be recognised that some flavonoids have biochemical effects that could translate into toxicities in humans, particularly if taken in large doses.10 11 A benefit to cardiovascular health is suggested by some epidemiological data but remains unproved in the absence of experimental studies of atherosclerosis or randomised clinical trials; indeed, the suspected benefits of even the classic dietary antioxidants remain to be demonstrated in clinical trials.12 The search for health promoting compounds will continue, but, given the complexities of our own biochemistry, the answers are not likely to be as simple as we would wish.

    We acknowledge the advice of Dr Jukka Salonen during the preparation of this paper.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    12. 12.