Re: Re-evaluation of the traditional diet-heart hypothesis: analysis of recovered data from Minnesota Coronary Experiment (1968-73)
Rather than applaud the efforts of Ramsden et al, Walter Willett dismisses their paper as largely useless and “irrelevant”. He writes:
"We have known for many years that the classical diet-heart is incomplete, including recognition of the importance of both N-3 fatty acids and N-6 dietary fatty acids..."
As history shows, the initial focus of dietary CHD recommendations was increased consumption of oils rich in the omega-6 fatty acid linoleic acid (LA). This polyunsaturated fatty acid was widely touted as beneficial due to its cholesterol-lowering effects. Widespread awareness of the benefits of n-3 fatty acids, specifically those from fish, did not occur until much later.
While Willett would have us believe soybean oil is a lifesaver, NIH researchers have linked America’s steadily worsening omega-6:omega-3 ratio over the last century to its embrace of this oil, the main constituent of which is LA. Estimated per capita consumption of soybean oil increased over 1000-fold from 1909 to 1999; as a result, the ratio of LA to ALA increased from 6.4 in 1909 to 10.0 in 1999. The researchers concluded this increased consumption of LA has likely decreased human tissue concentrations of the all-important n-3 fatty acids EPA and DHA[2].
Willett goes on to claim replacement of saturated fat with polyunsaturated fat “is almost certainly” a major factor, if not the primary factor, behind a decline in CHD mortality that began in the 1960s.
To accept this assertion, we need to accept that a worsening n-6:n-3 ratio is somehow good for coronary health. We also need to ignore the lack of decline in non-fatal CHD, because while CHD mortality did indeed decrease after the 1960s, total age-adjusted incidence of CHD did not[3-6]. If increased LA intake and decreased saturate intake were effective in preventing CHD, then they would surely have lowered both fatal and nonfatal CHD.
The real reason people continued to have just as many heart attacks but survived more of them is because of significant advances in extending the lives of CHD patients (including more emergency medical services, coronary-care units, anti-clotting drugs, etc) are far more coherent explanations for the decline in CHD deaths than consumption of LA.
In RCTs, the dietary fat changes recommended by Willett have been wholly ineffective in reducing CHD and overall mortality, as Ramsden et al note. Willett, however, protests that Ramsden et al "fail to mention that they earlier reported a benefit for incidence of coronary heart disease in a meta-analysis of randomized trials of trials in which saturated fat was replaced by a vegetable oils high in linoleic acid with a small amount of N-3 fatty acids, usually as soybean oil." [sic]
What Ramsden et al actually wrote in the meta-analysis to which Willett refers was "... pooled analysis of the four randomized controlled trials that increased n-3 PUFAs alongside n-6 LA showed reduced cardiovascular mortality."[8]
What a pooled statistical analysis of heterogeneous trials shows and what an intervention actually does is not the same thing. Let’s take a look at those four trials and see what actually transpired:
-In the Oslo Diet-Heart Study, the experimental group did indeed experience lower CHD incidence and cardiovascular mortality. But the control group was disadvantaged by more subjects aged 60 or older, more overweight men and, by the end of the study, almost twice as many heavy smokers[9,10].
These factors alone could easily explain the CHD outcomes. As for diet, replacement of animal fats with soy oil was only one of several dietary interventions employed. The intervention group was also instructed to increase fruit, vegetable and nut intake, and to eliminate consumption of trans fat-rich margarines. The diet group were also liberally supplied with sardines canned in cod liver oil. Unlike soya oil, sardines and cod liver are rich in long-chain omega-3 fatty acids; when tested in isolation, LC n-3 fatty acids have a far more impressive track record in CHD and mortality reduction than plant-based n-3s.
-In the MRC trial, coronary and overall mortality were the same in the soy oil and control groups[11].
-STARS also involved numerous dietary interventions: More fruits, vegetables and starchy complex carbohydrate foods, reduced consumption of processed foods (including "cookies, pastry, cakes"). While the dietary guidelines given to the patients called for "strictly limiting" intake of meat, fish and dairy products, dietary records showed 3-fold higher DHA intake among intervention subjects. DHA can only be obtained from animal foods, so either the dietary records were wrong (in which case we have to doubt the remainder of the recorded intakes) or the intervention group ate more n-3-rich animal foods such as fish.
Also of key importance is the intervention group maintained lower caloric intake and as result lost weight (-2.9 kg), while weight did not change in the control group (+0.3 kg)[12,13].
Given the numerous potential confounders (weight loss, reduced processed food intake, higher animal n-3 intake) intake, it again constitutes poor science to pronounce an increase in the polyunsaturated:saturate ratio as the reason for beneficial mortality outcomes.
- The LA Veterans Administration Study observed reduced CHD deaths among those eating the polyunsaturated-rich diet. However, a significant increase in cancer deaths among the intervention group entirely negated the reduction in CHD mortality[14].
When autopsies were performed on deceased subjects the researchers found little difference in degree of atherosclerosis between the two groups. If anything, those in the soybean oil diet group, despite having lower serum cholesterol levels, had slightly more plaque build-up in the aorta[15].
There were more non-smokers in the intervention group and a significantly higher number of heavy smokers among the controls. Research has shown arteries of smokers are far more susceptible to undergo blood-stopping spasm, even when coronary angiography findings are normal[16]. This could easily explain the lower incidence of CHD among the intervention group.
Cancer incidence, in contrast, was highest in the intervention group. In animal studies, linoleate-rich oils consistently increase tumour incidence and growth[17].
The RCT evidence does not support, and in fact flatly refutes, Willett's assertion that linoleate-rich oils like soy impart heart-healthy, life-extending benefits.
References
1. Page IH, et al. Dietary fat and its relation to heart attacks and strokes. Circulation, 1961; 23:133-136.
2. Blasbalg TL, et al. Changes in consumption of omega-3 and omega-6 fatty acids in the United States during the 20th century. American Journal of Clinical Nutrition. First published ahead of print March 2, 2011 as doi: 10.3945/ajcn.110.006643.
3. Rosamond WD, et al. Trends in incidence of myocardial infarction and in mortality due to coronary heart disease, 1987 to 1994. New England Journal of Medicine,1998; 339: 861-867.
4. Center for Disease Control. Hospitalization Rates for Ischemic Heart Disease - United States, 1970-1986. MMWR Weekly, Apr 28, 1989; 38 (16); 275-276, 281-284.
5. Lampe FC, et al. Trends in rates of different forms of diagnosed coronary heart disease, 1978 to 2000: prospective, population-based study of British men. BMJ, 2005; 330: 1046.
6. Sytkowski PA, et al. Changes in risk factors and the decline in mortality from cardiovascular disease. The Framingham Study. New England Journal of Medicine, 1990; 322: 1635-1641.
7. Ramsden CE, 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.
8. Leren P. The effect of plasma cholesterol lowering diet in male survivors of myocardial infarction. A controlled clinical trial. Acta Medica Scandanavica Supplement, 1966; 466: 1-92.
9. Leren P. The Oslo Diet-Heart Study: Eleven Year Report. Circulation, Nov 1970; Vol. 42: 935-942.
10. Medical Research Council. Controlled trial of soya-bean oil in myocardial infarction. Lancet, 1968; 2: 693-699.
11. Watts GF, et al. Effects on coronary artery disease of lipid-lowering diet, or diet plus cholestyramine, in the St Thomas’ Atherosclerosis Regression Study (STARS). Lancet, 1992; 339: 563-569.
12. Watts GF, et al. Dietary fatty acids and progression of coronary artery disease in men. American Journal of Clinical Nutrition, 1996; 64: 202-209.
13. Dayton S, et al. Controlled trial of a diet high in unsaturated fat for prevention of atherosclerotic complications. Lancet, 1968; 2: 1060-1062.
14. Dayton S, et al. A controlled clinical trial of a diet high in unsaturated fat in preventing complications of atherosclerosis. Circulation, 1969; 40 (Suppl. II): 1-63.
15. Nitenberg A, et al. Acetylcholine-induced coronary vasoconstriction in young, heavy smokers with normal coronary arteriographic findings. American Journal of Medicine, 1993; 95: 71–77.
16. Ip C, et al. Requirement of essential fatty acid for mammary tumorigenesis in the rat. Cancer Research, 1985; Vol. 45, Issue 5: 1997-2001.
Rapid Response:
Re: Re-evaluation of the traditional diet-heart hypothesis: analysis of recovered data from Minnesota Coronary Experiment (1968-73)
Rather than applaud the efforts of Ramsden et al, Walter Willett dismisses their paper as largely useless and “irrelevant”. He writes:
"We have known for many years that the classical diet-heart is incomplete, including recognition of the importance of both N-3 fatty acids and N-6 dietary fatty acids..."
As history shows, the initial focus of dietary CHD recommendations was increased consumption of oils rich in the omega-6 fatty acid linoleic acid (LA). This polyunsaturated fatty acid was widely touted as beneficial due to its cholesterol-lowering effects. Widespread awareness of the benefits of n-3 fatty acids, specifically those from fish, did not occur until much later.
While Willett would have us believe soybean oil is a lifesaver, NIH researchers have linked America’s steadily worsening omega-6:omega-3 ratio over the last century to its embrace of this oil, the main constituent of which is LA. Estimated per capita consumption of soybean oil increased over 1000-fold from 1909 to 1999; as a result, the ratio of LA to ALA increased from 6.4 in 1909 to 10.0 in 1999. The researchers concluded this increased consumption of LA has likely decreased human tissue concentrations of the all-important n-3 fatty acids EPA and DHA[2].
Willett goes on to claim replacement of saturated fat with polyunsaturated fat “is almost certainly” a major factor, if not the primary factor, behind a decline in CHD mortality that began in the 1960s.
To accept this assertion, we need to accept that a worsening n-6:n-3 ratio is somehow good for coronary health. We also need to ignore the lack of decline in non-fatal CHD, because while CHD mortality did indeed decrease after the 1960s, total age-adjusted incidence of CHD did not[3-6]. If increased LA intake and decreased saturate intake were effective in preventing CHD, then they would surely have lowered both fatal and nonfatal CHD.
The real reason people continued to have just as many heart attacks but survived more of them is because of significant advances in extending the lives of CHD patients (including more emergency medical services, coronary-care units, anti-clotting drugs, etc) are far more coherent explanations for the decline in CHD deaths than consumption of LA.
In RCTs, the dietary fat changes recommended by Willett have been wholly ineffective in reducing CHD and overall mortality, as Ramsden et al note. Willett, however, protests that Ramsden et al "fail to mention that they earlier reported a benefit for incidence of coronary heart disease in a meta-analysis of randomized trials of trials in which saturated fat was replaced by a vegetable oils high in linoleic acid with a small amount of N-3 fatty acids, usually as soybean oil." [sic]
What Ramsden et al actually wrote in the meta-analysis to which Willett refers was "... pooled analysis of the four randomized controlled trials that increased n-3 PUFAs alongside n-6 LA showed reduced cardiovascular mortality."[8]
What a pooled statistical analysis of heterogeneous trials shows and what an intervention actually does is not the same thing. Let’s take a look at those four trials and see what actually transpired:
-In the Oslo Diet-Heart Study, the experimental group did indeed experience lower CHD incidence and cardiovascular mortality. But the control group was disadvantaged by more subjects aged 60 or older, more overweight men and, by the end of the study, almost twice as many heavy smokers[9,10].
These factors alone could easily explain the CHD outcomes. As for diet, replacement of animal fats with soy oil was only one of several dietary interventions employed. The intervention group was also instructed to increase fruit, vegetable and nut intake, and to eliminate consumption of trans fat-rich margarines. The diet group were also liberally supplied with sardines canned in cod liver oil. Unlike soya oil, sardines and cod liver are rich in long-chain omega-3 fatty acids; when tested in isolation, LC n-3 fatty acids have a far more impressive track record in CHD and mortality reduction than plant-based n-3s.
-In the MRC trial, coronary and overall mortality were the same in the soy oil and control groups[11].
-STARS also involved numerous dietary interventions: More fruits, vegetables and starchy complex carbohydrate foods, reduced consumption of processed foods (including "cookies, pastry, cakes"). While the dietary guidelines given to the patients called for "strictly limiting" intake of meat, fish and dairy products, dietary records showed 3-fold higher DHA intake among intervention subjects. DHA can only be obtained from animal foods, so either the dietary records were wrong (in which case we have to doubt the remainder of the recorded intakes) or the intervention group ate more n-3-rich animal foods such as fish.
Also of key importance is the intervention group maintained lower caloric intake and as result lost weight (-2.9 kg), while weight did not change in the control group (+0.3 kg)[12,13].
Given the numerous potential confounders (weight loss, reduced processed food intake, higher animal n-3 intake) intake, it again constitutes poor science to pronounce an increase in the polyunsaturated:saturate ratio as the reason for beneficial mortality outcomes.
- The LA Veterans Administration Study observed reduced CHD deaths among those eating the polyunsaturated-rich diet. However, a significant increase in cancer deaths among the intervention group entirely negated the reduction in CHD mortality[14].
When autopsies were performed on deceased subjects the researchers found little difference in degree of atherosclerosis between the two groups. If anything, those in the soybean oil diet group, despite having lower serum cholesterol levels, had slightly more plaque build-up in the aorta[15].
There were more non-smokers in the intervention group and a significantly higher number of heavy smokers among the controls. Research has shown arteries of smokers are far more susceptible to undergo blood-stopping spasm, even when coronary angiography findings are normal[16]. This could easily explain the lower incidence of CHD among the intervention group.
Cancer incidence, in contrast, was highest in the intervention group. In animal studies, linoleate-rich oils consistently increase tumour incidence and growth[17].
The RCT evidence does not support, and in fact flatly refutes, Willett's assertion that linoleate-rich oils like soy impart heart-healthy, life-extending benefits.
References
1. Page IH, et al. Dietary fat and its relation to heart attacks and strokes. Circulation, 1961; 23:133-136.
2. Blasbalg TL, et al. Changes in consumption of omega-3 and omega-6 fatty acids in the United States during the 20th century. American Journal of Clinical Nutrition. First published ahead of print March 2, 2011 as doi: 10.3945/ajcn.110.006643.
3. Rosamond WD, et al. Trends in incidence of myocardial infarction and in mortality due to coronary heart disease, 1987 to 1994. New England Journal of Medicine,1998; 339: 861-867.
4. Center for Disease Control. Hospitalization Rates for Ischemic Heart Disease - United States, 1970-1986. MMWR Weekly, Apr 28, 1989; 38 (16); 275-276, 281-284.
5. Lampe FC, et al. Trends in rates of different forms of diagnosed coronary heart disease, 1978 to 2000: prospective, population-based study of British men. BMJ, 2005; 330: 1046.
6. Sytkowski PA, et al. Changes in risk factors and the decline in mortality from cardiovascular disease. The Framingham Study. New England Journal of Medicine, 1990; 322: 1635-1641.
7. Ramsden CE, 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.
8. Leren P. The effect of plasma cholesterol lowering diet in male survivors of myocardial infarction. A controlled clinical trial. Acta Medica Scandanavica Supplement, 1966; 466: 1-92.
9. Leren P. The Oslo Diet-Heart Study: Eleven Year Report. Circulation, Nov 1970; Vol. 42: 935-942.
10. Medical Research Council. Controlled trial of soya-bean oil in myocardial infarction. Lancet, 1968; 2: 693-699.
11. Watts GF, et al. Effects on coronary artery disease of lipid-lowering diet, or diet plus cholestyramine, in the St Thomas’ Atherosclerosis Regression Study (STARS). Lancet, 1992; 339: 563-569.
12. Watts GF, et al. Dietary fatty acids and progression of coronary artery disease in men. American Journal of Clinical Nutrition, 1996; 64: 202-209.
13. Dayton S, et al. Controlled trial of a diet high in unsaturated fat for prevention of atherosclerotic complications. Lancet, 1968; 2: 1060-1062.
14. Dayton S, et al. A controlled clinical trial of a diet high in unsaturated fat in preventing complications of atherosclerosis. Circulation, 1969; 40 (Suppl. II): 1-63.
15. Nitenberg A, et al. Acetylcholine-induced coronary vasoconstriction in young, heavy smokers with normal coronary arteriographic findings. American Journal of Medicine, 1993; 95: 71–77.
16. Ip C, et al. Requirement of essential fatty acid for mammary tumorigenesis in the rat. Cancer Research, 1985; Vol. 45, Issue 5: 1997-2001.
Competing interests: No competing interests