The theory that dietary lipids including cholesterol and saturated fat are causative factors in the development of atherosclerosis is a widely held belief. This viewpoint is based on supposed evidence from the scientific literature. However, studies used as evidence for this dietary lipid atherosclerosis link are generally outdated and suffer from serious methodological flaws. In addition much of the research has been performed in metabolic ward trials where macronutrient manipulation is unnatural, or using animal models. The value of the information they convey is therefore limited and of questionable relevance to free living human populations. A number of tricks have been used by proponents of the dietary lipid theory of cardiovascular disease. Two favourites are the comparison of saturated fats with polyunsaturated fats, or inclusion of hydrogenated fats and saturated fats in a single category. This creates uncertainty as to the cause of the changes to lipid levels which allows the interpretation of the data to be manipulated.
A great example of such manipulation can be found in a paper published in the American Journal of Clinical Nutrition in 19841. In this study, pigs were used as a model of human atherosclerosis and their dietary lipids manipulated to induce atherosclerosis. Weaning pigs were fed high fat diets for 6 months containing either high oleic safflower oil, lard, or a partially hydrogenated soybean oil. Analysis of the fatty acid composition of the adipose tissues of the pigs revealed that the triglycerides in adipocytes reflected the composition of the dietary fatty acid intake. However, the fatty acid composition of the heart did not as closely reflect the dietary intake, suggesting that the heart tissue preferentially accumulates specific fatty acids and rejects others. Both the partially hydrogenated soybean oil and the lard caused decreases in prostacyclin concentration in artery walls. Prostacyclin inhibits platelet activation and is an effective vasodilator. Based on this, both lard and hydrogenated soybean oil may have prothrombotic effects.
Excluding the fact that pigs are not humans and that animal data is not extrapolatable to humans, this study was problematic for a number of reasons. In particular, the authors made several conclusions from this data including the fact that the degree of fatty acid saturation was a contributory factor in the development of atherosclerosis and that hydrogenated fatty acids were not involved. However, closer examination of the data reveals that it is not possible to make these statements based on the data presented. Firstly, lard is not a saturated fat because it contains more unsaturated than saturated fatty acids. The effect of lard on serum total cholesterol concentrations cannot therefore be attributed to its saturated fat content. Further, the use of safflower oil is controversial because it contains γ-linolenic acid (GLA, C18:2 (n-6)) that is cardioprotective and plant sterols that are beneficial to plasma lipid levels. That this oil raised plasma levels of triglycerides suggests it was rancid and therefore should not have been used in the study.
Further, while hydrogenation does increase the degree of saturation of an oil it also increases the degree of trans fatty acids contained within the oil. This was reflected in the high concentration of trans-monounsaturated fatty acids in the soybean oil. However, to conclude that it was the degree of saturation, rather than the presence of the trans fatty acids that were the cause of reduced prostacyclin concentrations is not therefore possible. The same can also be said for lard. While lard does contain saturated fat, it is richest in unsaturated fat. The ability of lard to decrease prostacyclin concentrations in arterial tissue is therefore not attributable to the saturated fat content. The methodological problems with this study are extensive, because the number of variables between treatments does not allow accurate conclusions to be draw. In this respect the study is a mess, and it is a wonder how such unscientific thinking was able to pass the peer review process. Care when reading such studies is important as the conclusions drawn often to not reflect the data.
RdB
