EDITORIAL REVIEW Lipids: how do we convert experimental data to policy? Richard J. Deckelbaum a and Philip C. Calder b a Institute of Human Nutrition and Department of Pediatrics, Columbia University, New York, New York, USA; and b Institute of Human Nutrition, University of Southampton, Southampton, UK Correspondence to Richard J. Deckelbaum, MD, Institute of Human Nutrition, Columbia University, 630 West 168th Street, PH 15-1512, New York, NY 10032, USA Tel: +1 212 305 4808; fax: +1 212 305 3079; e-mail: rjd20@columbia.edu Current Opinion in Clinical Nutrition and Metabolic Care 2005, 8:113–114 Abbreviations DHA docosahexaenoic acid EPA eicosapentaenoic acid # 2005 Lippincott Williams & Wilkins 1363-1950 Over the past three or four decades, we have seen substantial changes in recommendations as to how the consumption of dietary fats can be altered to promote health and prevent disease. Early on, the major culprit linking heart disease to diet was cholesterol. Later, we learned that saturated fats were far more potent at increasing blood cholesterol levels, particularly LDL- cholesterol, than dietary cholesterol itself. Replacing saturated fats with high amounts of polyunsaturated fats in the diet seemed a sensible recommendation, and was adopted by many groups providing dietary guidelines. Promotion of the intake of polyunsaturated fatty acids was largely focused upon the n-6 family, with a resultant increased consumption of linoleic acid. However, more recently, there has been an increased awareness of the potential adverse effects of high intakes of n-6 polyun- saturated fatty acids, associated with their conversion to pro-inflammatory eicosanoids, and their ability to act as pro-oxidants. Concern with oxidant stress and the ability to decrease this led to large numbers of studies showing the antiox- idant potentials of vitamin C, beta carotene, and vitamin E, among others. Large clinical trials with single antiox- idants, or combinations, at ‘therapeutic’ doses were con- ducted. However, beta carotene at high levels of intake led to an increased risk of lung cancer in smokers [1]. Multiple antioxidant supplementation (vitamin E, vitamin C, beta carotene, selenium), added to drug regi- mens for cholesterol lowering, surprisingly decreased the benefit of drug therapy alone on cardiovascular disease- related endpoints [2,3]. Most recently, data have been released indicating that high levels of vitamin E intake might be linked to increases in overall mortality [4]. Clearly, molecules that were thought to act through a single pathway in improving biological outcomes can also act through a myriad of other mechanisms. In this issue, the article of Williams and Fisher (pp. 139–146) reviews the balances between pro and anti-oxidant effects and stresses the importance in some settings of oxidant stress. In the article by von Eckardstein and colleagues (pp. 147–152), the pleuripotential roles of HDL in pathways other than reverse cholesterol transport are reviewed. These are but two examples of complex intertwining pathways that can affect cardio- vascular, as well as many other, diseases. Another area of increasing interest is the potential for long-chain n-3 fatty acids, such as those found in oily fish and fish oils, to modulate biological pathways relating to positive health outcomes such as the prevention of car- diovascular disease, decreasing inflammatory responses, and improving mental performance in infants and the elderly. At the same time, data are emerging on the dangers of some potential toxins, such as mercury and dioxin-like compounds, carried in fish and potentially in fish oils. This concern led to recommendations to restrict the fish intake of women before conception and during pregnancy. Nevertheless, the potential benefit of long- chain n-3 fatty acids is high. However, there is confusion and contradiction in many areas of the literature regard- ing studies with n-3 fatty acids. This partly relates to the different approaches for supplying these fatty acids that have been used: these include increasing fish intake, providing the essential precursor for eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) as a-linolenic acid, or providing EPA and DHA themselves. However, the amounts of EPA and DHA in fish oils differ, as do the ratios between them. In addition, studies have used different doses of n-3 fatty acids and have been of different durations. Finally, studies have frequently used heterogeneous groups of subjects/patients, thus potentially masking effects that may be evident in only a subset (e.g. those with very low EPA or DHA status or with particular clinical manifestations). These variations undoubtedly contribute to the variability in outcomes, and can partly explain the disparate results in some of the trials discussed by Decsi and Koletzko (pp. 161 – 166) and by Cawood and colleagues (pp. 153–159) in this issue. Clearly, there are many considerations before we can convert experimental data to nutrition policy. With 113