Please enjoy this post from the archives dated November 8, 2013
By Gregory Katz, MD
Faculty Peer Reviewed
Everyday in clinic, we tell our patients to choose foods low in saturated fat. Because these foods raise plasma cholesterol, the thinking goes, they cause heart disease. Today, every major medical organization – from the American Heart Association to the Harvard School of Public Health to the USDA [1-3] – recommends a diet low in saturated fat to prevent and treat heart disease. The fat-cholesterol-heart disease connection is so thoroughly integrated into both medicine and popular culture that it’s become dogma. But since the initial data on saturated fat and cholesterol were published, our understanding of heart disease and cholesterol has significantly improved. We’ve moved from looking at total cholesterol to measuring LDL, HDL, and triglycerides. And as we better understand the pathophysiology of vascular disease, we may start to move away from cholesterol and directly measure lipoproteins instead. But even as our risk factor assessment has evolved, the diets we recommend to our patients have not. Are we due for a more nuanced approach to evidence-based dietary recommendations? Let’s take a look at the evidence.
The Early Observational Research
The first hypothesis that saturated fat might cause heart disease was developed more than a half century ago from data on the eating patterns of various nations. Back in the 1950s, researchers noted the epidemiologic relationship between heart disease and saturated fat consumption.[5,6,7] These associations were reinforced by data from the Framingham Heart Study which showed an association between hyperlipidemia and heart disease. Critics at the time noted flaws in the original epidemiologic research, pointing out that additional available data would eliminate the saturated fat-heart disease association. And while the Framingham study did find that cardiovascular disease was five times more likely for men with cholesterol over 260 than under 200, there was no difference between these groups in the type or quantity of fat consumed. The investigators also noted that cholesterol has “no predictive value” for women. And data on plasma cholesterol doesn’t necessarily tell us much about the effect on heart disease. A 1969 report from the National Heart, Lung, and Blood Institute stated, “It is not known whether dietary manipulation has any effect whatsoever on coronary heart disease.”
Metabolic Ward Data
Dietary research is notoriously hard to carry out. Researchers often rely on food frequency questionnaires to determine what participants in the study ate. Trusting their data requires placing faith in the ability of patients to recall exactly how frequently they eat various types of food. Many researchers believe these questionnaires to be a totally unreliable method of quantifying what people actually eat. Investigations on institutionalized patients provide a way around this data collection conundrum. The Finnish Mental Hospital Trial and the Los Angeles Veterans Adminstration Study are the two most well known papers in this area. The Finnish trial was a crossover study looking at patients in two psychiatric hospitals in Helsinki; one hospital served its patients full fat milk and butter while the other served unsaturated vegetable oils and filled milk (milk that has been filtered and had its fat replaced by emulsified vegetable oil). At the end of 6 years, the hospitals switched diets. The Finnish study found a 50% relative risk reduction in cardiovascular mortality (although no change in total mortality) in male patients fed the low saturated fat diet. Critics of the Finnish study have noted that the study was poorly controlled: almost half of the participants either entered or left the study over its 12 year duration and sugar consumption varied by more than 50% over the course of the trial. The Los Angeles VA study found an 18% relative risk reduction in cardiovascular deaths by replacing animal fat with vegetable oil, but no difference in overall mortality.
Newer research and confounding observations
More recent investigations have only served to complicate this issue further. After a 2008 expert meeting, the Food and Agricultural Organization of the WHO concluded that “Insufficient evidence relating to effect on the risk of heart disease in replacing saturated fat with carbohydrates” and “Probable evidence that replacing saturated fats with refined carbohydrates may increase risk of heart disease and favor metabolic syndrome development.” The Women’s Health Initiative found that a reduction in saturated fat did not significantly reduce the risk of heart disease or stroke. The A to Z weight loss study published in JAMA found better blood pressure and lipid profiles in patients following the Atkins diet compared to those reducing their saturated fat intake. A 2010 meta-analysis of prospective cohort studies looking at saturated fat and heart disease concluded, “There is no significant evidence for concluding that dietary saturated fat is associated with an increase in CVD.”
While many of the studies investigating the effect of diet on health look at proxies for cardiovascular disease like cholesterol levels and blood pressure instead of measuring hard outcomes like cardiac events and mortality, others confound the effect of dietary changes by measuring the effects of multiple simultaneous interventions. The most famous example of this is Dr. Dean Ornish, who has demonstrated an incredible impact on cardiovascular disease through his research on comprehensive lifestyle changes. Dr. Ornish attributes the effect to his ultra low fat diet, but his is still a hypothesis waiting to be rigorously tested.
Reconciling Discordant Observations: The Vital Importance of Replacement Nutrients
Caveats abound in the research on diet and heart disease and we must take caution before drawing conclusions about healthy eating. When counseling our patients, it is vital to remember that decreasing saturated fat changes two variables in the food consumption equation: the removal of saturated fat and the addition of something else to replace it. Indeed, the replacement that our patients choose is likely to be the factor that determines whether a reduction in saturated fat consumption is helpful or harmful.
A look at some other populations provides useful context. When the Tokelau people, Pacific Islanders who consume more than half of their calories from saturated fat, migrate to New Zealand, their saturated fat consumption goes down by half. But their rates of heart attack go up. The Masai, an African tribe, are another excellent example. When they are young, they eat a diet almost exclusively composed of blood, milk, and meat, at least one third of calories from saturated fat. As they get older, the Masai that continue this diet are remarkably free of atherosclerosis, even after age 60. But those Masai that reduce their consumption of these substances high in saturated fat and replace them with flour, sugar, and shortening begin to develop atherosclerosis.
Telling patients to reduce saturated fat consumption without giving them instructions on what to replace it with is a gamble. While the evidence suggests that replacing saturated fat with unsaturated fat reduces the risk of cardiovascular disease, it appears that replacing saturated fat with carbohydrates tends to worsen insulin sensitivity and lead to development of metabolic syndrome.[23,24,25].
A 2011 Cochrane Review found that replacing saturated fat with unsaturated fat can reduce relative risk of cardiovascular events by 14%, which equates to a 1% absolute risk reduction over two years. This is a modest, albeit significant benefit for a major lifestyle change. So we should be open with patients about the magnitude of risk reduction from adopting our proposed dietary recommendations. And we should also be careful about how we frame the changes, being honest about the risk they face by substituting carbohydrates for saturated fat.
This absolute risk reduction is one of the most important take home points when thinking about evidence based dietary recommendations. Smoking cessation and blood pressure control are likely to be lifestyle interventions with a greater magnitude of benefit than saturated fat reduction. We should allocate time in our clinical practices to discuss lifestyle modification, but we should also be mindful of the absolute benefits that we hope to attain.
Dr. Gregory Katz is a 2nd year resident at NYU School of Medicine
Peer reviewed by Robert Donnino, MD, Cardiology Section Editor, Clinical Correlations
Image courtesy of Wikimedia Commons
5. Keys A. Atherosclerosis: A problem in newer public health. Mount Sinai Hosp NY, 20 (1953), pp. 118–139 http://www.ncbi.nlm.nih.gov/pubmed/13085148
6. Keys A. Coronary heart disease in seven countries. Circulation, 41 (Suppl. 1) (1970), pp. 1–211
7. A. Keys, A. Menotti, M.J. Karvonen et al. The diet and 15-year death rate in The Seven Countries Study. Am J Epidemiol, 124 (1986), pp. 903–915
8. Kannel WB, Gordon T. The Framingham diet study: Diet and the regulation of serum cholesterol. The Framingham study. An epidemiologic investigation of cardiovascular disease. Section 24, Washington, DC, 1970.
9. J. Yerushalmy, H.E. Hilleboe. Fat in the diet and mortality from heart disease. A methodologic note. NY State J Med, 57 (1957), pp. 2343–2354
10. Taubes G. The Soft Science of Dietary Fat. Science 30 March 2001. Vol 291.
11. Schaefer E et al. Lack of efficacy of a food frequency questionairre in assessing dietary macronutrient intakes in subjects consuming diets of known composition. Am J Clinical Nutrition. March 2000;71(3):746-751 http://ajcn.nutrition.org/content/71/3/746.long
12. Miettinen M et al. Effect of cholesterol lowering diet on mortality from coronary disease and other causes. A twelve year trial in men and women. Lancet. 1972 Oct 21;2(7782):835-8.
13. Dayton S et al. A controlled diet high in unsaturated fat in preventive complications of atherosclerosis. Circulation 1969;40(11):1-63
15. Howard BV, Van Horn L, et al. Low fat dietary pattern and risk of cardiovascular disease: Women’s Health Initiative randomized controlled dietary modification trial. JAMA 2006 Feb 8; 295(6):655-66
16. Gardener C, King A et al. Comparison of the Atkins, Zone, Ornish, and LEARN Diets for Change in Weight and Related Risk Factors Among Overweight Premenopausal WomenThe A TO Z Weight Loss Study: A Randomized Trial. JAMA. 2007;297(9):969-977. doi:10.1001/jama.297.9.969.
17. Siri-Tarino PW, Sun Q, Hu FB, Krauss RM. Saturated fat, carbohydrate, and cardiovascular disease. Am J Clin Nutr. 2010 Mar;91(3):502-9. Epub 2010 Jan 20.
18. Ornish D, Scherwitz LW, Billings JH, Brown SE, Gould KL, Merritt TA, Sparler S, Armstrong WT, Ports TA, Kirkeeide RL, Hogeboom C, Brand RJ. Intensive lifestyle changes for reversal of coronary heart disease. JAMA. 1998 Dec 16;280(23):2001-7. http://www.ncbi.nlm.nih.gov/pubmed/9863851
19. Stanhope JM, Sampson VM. The Tokelau Migrant Study. J. Chronic Disease. 1981;34(2-3):45-55.
20. Ho KJ et al. The Masai of East African: some unique biological characteristics. Arch Pathol. 1971 May;91(5):387-410
21. Biss K et al. The Masai’s protection against atherosclerosis. Pathologic Microbiology. 1970;35(1):198-204.
22. Mann GV. Atherosclerosis in the Masai. Am J Epidemiology. 1972 Jan;95(1):26-37 http://aje.oxfordjournals.org/content/95/1/26.short
23. Volek et al. Carbohydrate restriction has a more favorable impact on the metabolic syndrome than low fat diet. Lipids 2009;44(4):297-309
24. Siri PW, Krauss RM. Influence of dietary carbohydrate and fat on LDL and HDL particle distribution. Curr Atherosclerosis Rep 2005 Nov;7(6):455-9 http://www.ncbi.nlm.nih.gov/pubmed/16256003
25. Weinberg SL. The diet-heart hypothesis: a critique. J Am Coll Cardiology. 2004:43(5):731-733
26. Hooper L et al. Reduced or modified dietary fat for preventing cardiovascular disease. Cochrane Database System Review. 2011 July 6;(7):CD002137 http://www.ncbi.nlm.nih.gov/pubmed/11687015