Intended for healthcare professionals

CCBYNC Open access

Re-evaluation of the traditional diet-heart hypothesis: analysis of recovered data from Minnesota Coronary Experiment (1968-73)

BMJ 2016; 353 doi: (Published 12 April 2016) Cite this as: BMJ 2016;353:i1246

Re: Re-evaluation of the traditional diet-heart hypothesis: analysis of recovered data from Minnesota Coronary Experiment (1968-73)

The response to this newly disclosed evidence and new meta-analysis by those still promoting the replacement of saturated fat with polyunsaturated vegetable oil seems to be, that while the incidence of obesity and diabetes have increased because the population is not, overall, following the dietary guidelines, heart disease mortality has dropped because the population has, overall, followed the dietary guidelines.

But, as Anthony Colpo has pointed out in his recent rapid response, the incidence of coronary disease has stayed much the same, and the drop in mortality is mainly due to improvements in treatment.[1] Macrovascular disease attributable to diabetes, as measured by the rate of amputations, has increased exponentially in the interim; Professor Tim Noakes has recently drawn attention to the similarities between these pathologies. It cannot be said that our arteries are healthier than they have ever been.[2]

In the Framingham study, which may be the longest running and most extensive observational test of the lipid hypothesis, higher levels of LDL cholesterol were predictive of significant reductions in the risk of type 2 diabetes.[3] This is consistent with biomarker studies showing that a higher serum level of saturated fatty acids found in ruminant fats is associated with a large reduction in type 2 diabetes risk, and also consistent with the recent meta-analysis of observational studies showing that higher vs lower consumption of ruminant trans fats was associated with reduced type 2 diabetes risk.[4]

Are there indications that a higher intake of linoleic acid from oils and spreads may be harmful in specific conditions? Non-alcoholic liver disease (NAFLD) is predictive of, and plausibly causational in, both type 2 diabetes and cardiovascular disease. Two recent studies have showed the successful treatment of NAFLD with diet. In one, GGT, HbA1c and BMI were all reduced by a very low-carbohydrate dietary approach in which polyunsaturated vegetable oils were replaced with fat from dairy and olive oil.[5] In the other, biomarkers of omega 6 metabolism were found to be elevated. Sugar was restricted to under 10% of energy and linoleic acid was restricted to below 4% of energy, resolving NAFLD over a 6 month period.[6] In Poland, where this study took place, the recommended upper limit for linoleic acid intake is 3% of energy, showing that not all authorities are in agreement with the American epidemiologists; in particular, the lipid biochemists consulted by European government bodies tend to have different opinions on this matter.

Chronic hepatitis C virus (HCV) infection is associated with a reduction in cholesterol, but this confers no benefit in terms of cardiovascular disease. Linoleic acid lowers cholesterol by upregulating the LDL receptor; HCV circulates in association with triglyceride-rich lipoproteins and infects hepatocytes via the LDL receptor complex. In a study of a population with chronic HCV infection, a higher intake of polyunsaturated fat was associated with an increased incidence of steatosis, but a higher intake of saturated and monounsaturated fat was not.[7]

Professor Brunner in his response drew a comparison between statins and linoleic acid, but this comparison may not be safe. Statins lower cholesterol by HMG-CoA reductase inhibition and may be effective in the secondary prevention of heart attacks because they stabilize arterial plaques.[8] However, arterial plaques with a higher content of linoleic acid are less stable than plaques with a high content of saturated fatty acids.[9] Rather than inhibiting HMG-CoA reductase, linoleic acid increases its activity, as 22% of the carbon from linoleic acid is converted to cholesterol in the liver, despite LA also increasing the liver’s uptake of cholesterol via LDL receptor upregulation.[10]

It has long been known, through multiple animal experiments with highly significant and consistent results, that polyunsaturated fats are essential for the progression of alcoholic liver disease, which can be prevented by feeding a diet sufficiently low in polyunsaturated fat and high in saturated fats, which have separate protective effects via multiple mechanisms.[11] A potentially life-saving treatment in humans has never been trialed because of concerns about cardiovascular safety which may have been unfounded. Meanwhile our citizens, still drinking much more heavily than public health experts would like, are likely today to fill their stomachs with food laden with polyunsaturated vegetable oils.

When some experts say saturated fat is still harmful, they often refer to observational or feeding studies concerning the fat in muffins, pizzas, frankfurters, milkshakes, and other processed foods that are also sources of refined carbohydrates, for example in “dietary pattern” type analyses. When other experts say saturated fat is not harmful, they refer to studies directly concerning cheese and milk, steak and coconut. Are these really conflicting opinions?

Dietary saturated fat is in the position of someone wrongly accused and convicted on mistaken and falsified evidence in the middle of a moral panic. In such cases, although better evidence becomes available, other suspects appear, and cooler heads prevail, acquittal is always a slow and painful process.

Meanwhile, the accused is unable to make whatever contribution to society they have always been capable of.



[3] Andersson C, Lyass A, Larson MG, Robins SJ, Vasan RS. Low-density-lipoprotein cholesterol concentrations and risk of incident diabetes: epidemiological and genetic insights from the Framingham Heart Study. Diabetologia. 2015 Dec;58(12):2774-80. doi: 10.1007/s00125-015-3762-x. Epub 2015 Sep 26

[4] de Souza, RJ, Mente, A, Maroleanu et al. Intake of saturated and trans unsaturated fatty acids and risk of all cause mortality, cardiovascular disease, and type 2 diabetes: systematic review and meta-analysis of observational studies. BMJ 2015;351:h3978

[5] Unwin DJ, Cuthbertson DJ, Feinman R, Sprung VS. A pilot study to explore the role of a low-carbohydrate intervention to improve GGT levels and HbA1c. Diabesity in Practice. 2015;4:102–8.

[6] Maciejewska D, Ossowski P, Drozd A, et al. Metabolites of arachidonic acid in early stages of non-alcoholic fatty liver disease – a pilot study. Prostaglandins Other Lipid Mediat. 2015;121(B):184-9.

[7] Loguercio C, Federico A, Masarone M et al. The impact of diet on liver fibrosis and on response to interferon therapy in patients with HCV-related chronic hepatitis. Am J Gastroenterol. 2008 Dec;103(12):3159-66. doi: 10.1111/j.1572-0241.2008.02159.x. Epub 2008 Sep 11.


[9] Felton CV, Crook D, Davies MJ, Oliver MF. Dietary polyunsaturated fatty acids and composition of human aortic plaques. Lancet. 1994 Oct 29;344(8931):1195-6.

[10] Cunnane SC. Problems with essential fatty acids: time for a new paradigm? Prog Lipid Res. 2003 Nov;42(6):544-68.

[11] Nanji AA, Jokelainen K, Tipoe GL, Rahemtulla A, Dannenberg AJ. Dietary Saturated Fatty Acids Reverse Inflammatory and Fibrotic Changes in Rat Liver Despite Continued Ethanol Administration. JPET November 1, 2001 vol. 299 no. 2 638-644.

Competing interests: No competing interests

21 April 2016
George D Henderson
Research Associate
Human Potential Centre, Auckland University of Technology
17 Antares Place, Auckland, New Zealand