That coconut oil contributes to cardiovascular disease would appear noncontroversial because its saturated fat content increases plasma low-density lipoprotein (LDL) cholesterol concentration. Cholesterol-rich LDL is a major cause of atherosclerosis because it delivers its cholesterol load to the arterial wall and causes obstruction and inflammation. Nonetheless, coconut oil has been accorded much attention in the popular media as a potentially beneficial food product. In fact, a survey in 2016 found that 72% of Americans viewed coconut oil as a healthy food. This represents a remarkable success in marketing by the coconut oil and related industries calling coconut oil a natural, healthful product, despite its known action to increase LDL cholesterol, an established cause of atherosclerosis and cardiovascular events.
A systematic review, published in 2016, identified 7 trials that tested the effect of coconut oil on LDL cholesterol. In these trials, coconut oil was compared with oils that had a high content of unsaturated fats. Significant detrimental effects were found in 6 of them. The present study by Neelakantan, Seah, and van Damis an important advance over this systematic review in that it includes a total of 17 published trials, takes a quantitative rather than a descriptive approach, and includes a range of outcomes relevant to assessing cardiovascular and metabolic health. This meta-analysis found that coconut oil significantly increased plasma LDL cholesterol and high-density lipoprotein (HDL) cholesterol, and had no effect on triglycerides, body weight, body fat, and markers of glycemia and inflammation in comparison with nontropical vegetable oils. Overall, this meta-analysis is rigorously conducted and reported, putting the results in the context of cardiovascular disease prevention.
Coconut oil is composed mainly of the saturated fatty acid, lauric acid (12 carbon atoms), but also of other long-chain saturated fatty acids, myristic (14 carbon atoms) and palmitic acids (16 carbon atoms). Mensink performed a comprehensive systematic review with meta-regression of each of these fatty acids on plasma LDL cholesterol and other lipoproteins. Mensink's review considered all sources of lauric, myristic, and palmitic acids, not only from coconut oil, but also in other foods such as dairy fat, palm kernel, and palm oil. All these saturated fatty acids increased LDL cholesterol. Lauric acid, the most prevalent fatty acid in coconut oil, had a significant linear effect on LDL cholesterol. Mensink used carbohydrate as the direct comparator nutrient for the fatty acids. His approach found even more of an effect on LDL cholesterol of these saturated fatty acids in comparison with mono- and polyunsaturated fatty acids, combining the 2 estimates (coconut oil minus carbohydrate) + (carbohydrate minus unsaturated fats). This is a practical way to illustrate the dietary application of the present meta-analysis, because unsaturated oils like soybean, corn, olive, or peanut oils are practical replacements for coconut oil.
Lauric acid is often classified as a medium-chain fatty acid, lumped with shorter chain fatty acids that have 6, 8, or 10 carbons. However, lauric acid, with its 12-carbon atoms, acts biologically like a long-chain fatty acid absorbed by packaging into chylomicrons. This mechanism increases LDL cholesterol. True medium-chain fatty acids are absorbed directly into the portal circulation and do not affect LDL cholesterol. Coconut oil is not an oil that acts as if its main components are medium-chain fatty acids. Coconut oil has approximately 13% true medium-chain fatty acids having 6, 8, or 10 carbon atoms. Thus, classifying lauric acid as a medium-chain fatty acid is a misnomer, going against its biological action as a long-chain fatty acid. Neelakantan and colleagues wrote a well-reasoned section in the introduction that rebuts this argument, and stands by the well-established absorption of lauric acid to form chylomicrons, like other long-chain saturated fatty acids.
The database includes small numbers of trials that could be used in analyses of the effects on LDL cholesterol of specific dietary comparisons, such as coconut oil versus butter, or coconut oil versus individual nontropical vegetable oils. Although not the primary aim of the present study, these comparisons could be used to form a hierarchy of health effects of cooking oils. However, the effect on LDL cholesterol of additional dietary comparisons may be estimated well by the meta-regression analysis on the component fatty acids.
Although coconut oil increases plasma HDL cholesterol, it is impossible to know if this is a beneficial mechanism in cardiovascular disease. Although HDL cholesterol is a robust risk marker for cardiovascular disease, genetic studies and HDL-raising drugs have not so far supported a causal relationship between HDL cholesterol and cardiovascular disease. HDL, the lipoprotein, is composed of a huge array of subparticles that may have adverse or beneficial actions.[7,8] It is unknown which, if any, foods or nutrients that raise HDL cholesterol do so in a way that reduces atherosclerosis and coronary events. Thus, effects on cardiovascular disease of foods or nutrients cannot be judged from changes in HDL cholesterol.
There is no randomized clinical trial that determined the effect of coconut oil on cardiovascular events such as myocardial infarction, heart failure, or stroke. Such a trial is unlikely to be attempted because of the high cost of hundreds of millions of dollars, large numbers of participants, and many years of treatment with coconut oil and an appropriate control fat. The inevitable rise in LDL cholesterol sustained over years in those assigned to coconut oil will create an ethical concern of harm, and may stop the trial before a definitive result is obtained. This situation is relevant to much of nutrition research. This limitation can be countered with evidence from the effects of foods on established cardiovascular risk factors, such as LDL cholesterol, and on incident cardiovascular events in large prospective, observational cohorts.
Advertisements give the impression that purportedly beneficial constituents other than saturated fat compensate for its adverse effects on LDL cholesterol. Yet, controlled trials in humans are not available that support beneficial actions of the components of coconut oil on cardiovascular disease risk factors or mechanisms.
Coconut oil may be viewed as one of the most deleterious cooking oils that increases risk for cardiovascular disease. Even in comparison with palm oil, another tropical oil with high saturated fat content, coconut oil increased LDL cholesterol. Replacing coconut oil with nontropical unsaturated vegetable oils, especially those rich in polyunsaturated fat, will have a health benefit. We believe that the results from the present meta-analysis can inform the development of nutrition recommendations and US Department of Agriculture dietary guidelines. In culinary practice, coconut oil should not be used as a regular cooking oil, although it can be used sparingly for flavor or texture.
Circulation. 2020;141(10):815-817. © 2020 American Heart Association, Inc.