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)

Why the Minnesota Coronary Experiment matters

In our report of recovered documents and data from the Minnesota Coronary Experiment (MCE)[1]—we found that replacement of saturated fat (SFA) with vegetable oil rich in linoleic acid (LA) effectively lowered serum cholesterol, but did not translate to improved health outcomes. Paradoxically, MCE participants who had greater reduction in serum cholesterol had higher, rather than lower, risk of death. Moreover, in our systematic review and meta-analysis of randomized controlled trials (RCT) that specifically replaced SFA with vegetable oils rich in LA there was no indication of benefit. These findings were interpreted cautiously with an emphasis on limitations. The responses ranged from appreciative to dismissive. Now, after the dust has settled, we respond to some BMJ comments and provide additional context.

Dose matters
Comment: the MCE intervention “was never consumed by any appreciable number of Americans”-Walter Willett

The MCE intervention provided ≈13%E (%E) as LA[1]. According to NHANES, more than 13 million American adults currently consume ≥13%E as LA[2].
Comment: MCE results have “no relevance to current dietary recommendations”-Walter Willett

In 2016, the NutriCoDE group defined optimal LA intake as 12%E (±1.2)[3]—a range that overlaps with the MCE intervention. In 2009, the American Heart Association advised “at least 5 to 10%E” as LA[4].
Traditional and modern Mediterranean diets contain about 2%E as LA[5,6]. The average American currently consumes ≈7%E as LA[2,7]. These levels depend on widespread consumption of concentrated vegetable oils[7]. Individuals eating only minimally-processed whole foods (customary until about 100 years ago) would have consumed about 2-3%E as LA.
Thus, wide ranges of LA are currently advised and consumed; current intakes in non-Mediterranean populations are much higher than historic norms. A central question for current dietary recommendations is:
What is the optimal intake of LA?

• Should it be 2-3%E, consistent with all diets prior to industrial oil processing[7,8], and Mediterranean-style diets[5,6]?

• Or should it be closer to 7%E, average intake in the U.S.[2], and consistent with some current guidelines[4]?

• Or should it be closer to 12%E, recently defined as optimal for the entire world population[3]? To our knowledge, there is a lack of evidence supporting the safety or efficacy of 12%E as LA.

Other tissues and diseases matter
Biochemical consequences of high intake of LA extend far beyond lowering serum cholesterol and the cardiovascular system[9-11](reviewed in [1,20]). Emerging research shows that high LA intakes increase bioactive lipid mediators in many tissues ranging brain[12] to mammary tissue[13]. These non-cholesterol mediators are implicated in the pathogenesis of many diseases[1,9-11,14,15,20]. While more research is needed, there is a possibility for unintended harm. There is also very little biochemical or clinical data about the effects of high LA intakes in special populations, such as older adults, pregnant women, and young children. Since the effects of high LA diets are poorly understood and there is a possibility for harm, it is crucial to know if the proposed benefit of LA on cardiovascular health is supported by RCT evidence.

Randomization matters
In a detailed systematic review and meta-analysis of RCT that specifically replaced SFA with vegetable oils rich in LA, we found no indication of benefit. This is important because RCTs allow stronger causal inferences and are less susceptible to confounders than non-randomized studies. Findings were presented in the context of the surprising lack of causal RCT data on this topic, emphasizing key limitations. For the broader context needed to fully evaluate the results and limitations of our review and meta-analysis, please see our manuscript[1] and appendices
Below we summarize some of this published information in response to BMJ comments.
Comment: Our meta-analysis of RCTs “omitted the Finnish Mental Hospital trial, an important study”
-Bhupathiraju & Hu

The Finnish Mental Hospital Study (FMHS)16 was not suitable for inclusion because it was not an RCT (reviewed in[16,17]). The FMHS was also potentially confounded by unequal use of thioridazine (linked to electrocardiographic abnormalities and sudden death)[1,17].

Comment: “The Cochrane review of 15 well-designed randomized controlled trials concluded that reducing SFA reduces the risk of cardiovascular events by 17 percent)”-Liebman, Blackburn & Jacobson

The Cochrane review and meta-analysis of RCTs by Hooper et al18 was not specifically focused on replacing SFA with LA. As such, it included several trials that lowered total fat and SFA without increasing LA, and trials that provided large doses of omega-3 EPA+DHA.

In reply to the comment that “a previous meta-analysis, which included the MCE, found significant benefits of replacing SFA with polyunsaturated fats on both serum cholesterol and incidence of major cardiovascular events.” -Bhupathiraju & Hu

The cited meta-analysis by Mozaffarian et al[19] was not specifically focused on replacing SFA with LA. It included trials that provided large doses of EPA+DHA from seafood and cod liver oil, trials with minimal increases in LA, and trials with complex diet changes (e.g. sugar restriction, increase in soluble fiber). It also included the non-randomized FMHS, but not recently recovered RCT data from the Sydney Diet Heart Study (SDHS)[20]. Recovered SDHS data belatedly showed that replacement of SFA with vegetable oil rich in LA significantly increased the risk of death from coronary heart disease, despite lowering serum cholesterol.

Comment: MCE findings are not meaningful due to the “very short duration on the assigned diets”.-Walter Willett

The fact that “only about a quarter of randomized participants remained in the study for a year or longer” was emphasized as a key limitation in our manuscript[1]. Importantly, however, 2,355 MCE participants were exposed to MCE diets for ≥1 year. The original investigators emphasized this subsample, which alone is much larger than any other RCT testing the effects of replacing SFA with LA. There was no indication of benefit; and although we didn’t emphasize this in the manuscript, there was an unfavorable separation of the mortality curves after about one-year of corn oil exposure (see Fig. 5 in[1]). This separation was most evident in older adults (Fig. 1, below after references).

We stand by our conclusion that available RCT evidence does not provide support for the hypothesis that serum cholesterol-lowering effects of replacing SFA with vegetable oil rich in LA translate to lower risk of death from coronary heart disease or all causes.

1. Ramsden CE, Zamora D, Majchrzak-Hong S, et al. Re-evaluation of the traditional diet-heart hypothesis: analysis of recovered data from Minnesota Coronary Experiment (1968-73). BMJ. 2016;353:i1246. PubMed PMID: 27071971. Pubmed Central PMCID: 4836695.
2. U.S. Department of Agriculture, Agricultural Research Service. Nutrient Intakes from Food: Mean Amounts Consumed per Individual, by Gender and Age, What We Eat in America, National Health and Nutrition Examination Survey (NHANES 2011-2012). [Internet].
3. Wang Q, Afshin A, Yakoob MY, et al. Impact of Nonoptimal Intakes of Saturated, Polyunsaturated, and Trans Fat on Global Burdens of Coronary Heart Disease. J Am Heart Assoc. 2016 Jan;5(1). PubMed PMID: 26790695. Pubmed Central PMCID: 4859401.
4. Harris WS, Mozaffarian D, Rimm E, et al. Omega-6 fatty acids and risk for cardiovascular disease: a science advisory from the American Heart Association Nutrition Subcommittee of the Council on Nutrition, Physical Activity, and Metabolism; Council on Cardiovascular Nursing; and Council on Epidemiology and Prevention. Circulation. 2009 Feb 17;119(6):902-7. PubMed PMID: 19171857. Epub 2009/01/28. eng.
5. Ferro-Luzzi A, Sette S. The Mediterranean Diet: an attempt to define its present and past composition. Eur J Clin Nutr. 1989;43 Suppl 2:13-29. PubMed PMID: 2689161.
6. Kushi LH, Lenart EB, Willett WC. Health implications of Mediterranean diets in light of contemporary knowledge. 2. Meat, wine, fats, and oils. Am J Clin Nutr. 1995 Jun;61(6 Suppl):1416S-27S. PubMed PMID: 7754997.
7. Blasbalg TL, Hibbeln JR, Ramsden CE, Majchrzak SF, Rawlings RR. Changes in consumption of omega-3 and omega-6 fatty acids in the United States during the 20th century. Am J Clin Nutr. 2011 May;93(5):950-62. PubMed PMID: 21367944. Pubmed Central PMCID: 3076650.
8. Kuipers RS, Luxwolda MF, Dijck-Brouwer DA, et al. Estimated macronutrient and fatty acid intakes from an East African Paleolithic diet. Br J Nutr. 2010 Dec;104(11):1666-87. PubMed PMID: 20860883. Epub 2010/09/24. eng.
9. Vangaveti V, Baune BT, Kennedy RL. Hydroxyoctadecadienoic acids: novel regulators of macrophage differentiation and atherogenesis. Ther Adv Endocrinol Metab. 2010 Apr;1(2):51-60. PubMed PMID: 23148150. Pubmed Central PMCID: 3475286.
10. Ramsden CE, Ringel A, Majchrzak-Hong SF, et al. Dietary linoleic acid-induced alterations in pro- and anti-nociceptive lipid autacoids: Implications for idiopathic pain syndromes? Mol Pain. 2016;12. PubMed PMID: 27030719.
11. Feldstein AE, Lopez R, Tamimi TA, et al. Mass spectrometric profiling of oxidized lipid products in human nonalcoholic fatty liver disease and nonalcoholic steatohepatitis. J Lipid Res. 2010 Oct;51(10):3046-54. PubMed PMID: 20631297. Pubmed Central PMCID: 2936759.
12. Taha AY, Hennebelle M, Yang J, et al. Regulation of rat plasma and cerebral cortex oxylipin concentrations with increasing levels of dietary linoleic acid. Prostaglandins Leukot Essent Fatty Acids. 2016;Forthcoming.
13. Johnson JA, Blackburn ML, Bull AW, Welsch CW, Watson JT. Separation and quantitation of linoleic acid oxidation products in mammary gland tissue from mice fed low- and high-fat diets. Lipids. 1997 Apr;32(4):369-75. PubMed PMID: 9113624.
14. Kirpich IA, Feng W, Wang Y, et al. Ethanol and dietary unsaturated fat (corn oil/linoleic acid enriched) cause intestinal inflammation and impaired intestinal barrier defense in mice chronically fed alcohol. Alcohol. 2013 May;47(3):257-64. PubMed PMID: 23453163. Pubmed Central PMCID: 3617059.
15. Liu H, Beier JI, Arteel GE, et al. Transient receptor potential vanilloid 1 gene deficiency ameliorates hepatic injury in a mouse model of chronic binge alcohol-induced alcoholic liver disease. Am J Pathol. 2015 Jan;185(1):43-54. PubMed PMID: 25447051. Pubmed Central PMCID: 4278357.
16. Turpeinen O, Karvonen MJ, Pekkarinen M, Miettinen M, Elosuo R, Paavilainen E. Dietary prevention of coronary heart disease: the Finnish Mental Hospital Study. Int J Epidemiol. 1979 Jun;8(2):99-118. PubMed PMID: 393644. Epub 1979/06/01. eng.
17. Ramsden CE, Hibbeln JR, Majchrzak SF, Davis JM. n-6 fatty acid-specific and mixed polyunsaturate dietary interventions have different effects on CHD risk: a meta-analysis of randomised controlled trials. Br J Nutr. 2010 Dec;104(11):1586-600. PubMed PMID: 21118617. Epub 2010/12/02. eng.
18. Hooper L, Martin N, Abdelhamid A, Davey Smith G. Reduction in saturated fat intake for cardiovascular disease. Cochrane Database Syst Rev. 2015;6:CD011737. PubMed PMID: 26068959.
19. Mozaffarian D, Micha R, Wallace S. Effects on Coronary Heart Disease of Increasing Polyunsaturated Fat in Place of Saturated Fat: A Systematic Review and Meta-Analysis of Randomized Controlled Trials. Plos Medicine. 2010 Mar;7(3):-. PubMed PMID: ISI:000276311600013. English.
20. Ramsden CE, Zamora D, Leelarthaepin B, et al. Use of dietary linoleic acid for secondary prevention of coronary heart disease and death: evaluation of recovered data from the Sydney Diet Heart Study and updated meta-analysis. BMJ. 2013;346:e8707. PubMed PMID: 23386268.

Fig 1. Risk of death from any cause by diet assignment in the prespecified subgroup of adults 65 and older (Broste Thesis1)
The unfavorable separation of the mortality curve for the corn oil group began approximately one-year after diet exposure.

Competing interests: No competing interests

26 May 2016
Christopher E Ramsden
Medical Investigator
Daisy Zamora, Sharon Majchrzak-Hong, Keturah R Faurot, Steven K Broste, Robert P Frantz, John M Davis, Amit Ringel, Chirayath M Suchindran, Joseph R Hibbeln
Section on Nutritional Neurosciences, Laboratory of Membrane Biochemistry and Biophysics, National Institute on Alcohol Abuse and Alcoholism, National Institutes of Health
Bethesda, MD, USA