Archives of Internal Medicine, Volume 152, July 1992, pp. 1371-2


On the possibility of a nut …

The findings reported by Fraser et al1 from the Adventist Health Study revive our interest in looking for data from prospective studies that show diet factors associated with favorable blood cholesterol or lipoprotein levels in free-living pop­ulations eventually lead to lower rates of coronary heart disease (CHD).  Most of what we know about the effects of diet factors, particularly the saturation of fat and cholesterol, on serum lipid parameters derives from metabolic ward-type studies.2,3 Alas, such findings, within a cohort studied over time have been disappointing, indeed the findings have been contradictory.  For example, in Framingham, Mass, the more satu­rated fat one ate, the more choles­terol one ate, the more calories one ate, the lower the person's serum choles­terol.  The opposite of what one saw in the 26 metabolic ward studies, the opposite of what the equations pro­vided by Hegsted et al2 and Keys et al3 would predict.  Only the interna­tional comparisons showed that the world could be lined up on choles­terol intake or saturated fat intake, and it would correlate with the rate of CHD.4  Of course, since these coun­tries differed in many other ways, the possibility that some unidenti­fied factor might explain the rate of CHD, loomed in one's thoughts. Eventually, diet intervention trials were done, and where the follow-up got out beyond 3 years, they all show the same thing. The larger the per­centage fall in cholesterol, the larger the percentage fall in CHD.5

In view of this, this study fails to describe a relationship of those tra­ditional dietary constituents, satu­rated fat and cholesterol, known to have an adverse effect on blood lip-ids, and thereby, on the subsequent development of coronary disease end points.  Only the Western Elec­tric study6 has shown dietary choles­terol to be related to the later devel­opment of CHD in a population study.  However, the authors of this Adventist study did show a slight increase in definite nonfatal myocardial infarction with eating cheese one to two times per week (RR = 1.97; 95% confidence interval, 1.27 to 3.04) and, in men, a relationship of eating beef to fatal CHD.  Whole wheat bread, thank goodness, lowered the nonfatal coronary disease rate.

The big finding was nuts. Nuts, eaten five or more times a week, ap­parently independently (Cox pro­portional hazard model analysis ad­justed by age, sex, smoking, exercise, relative weight, and high blood pressure) lowered the coro­nary fatal and nonfatal end points in half. But these are the Seventh-Day Adventists who already have a sev­enth of our heart attack rate, who live 7 years longer than we do. How could you cut this rate even lower, in half? Is this the first article showing a dramatic fall in coronary disease rates in men and women who are al­ready at low risk?

The first reaction of a population watcher is that there just has to be some other factor related to nut ingestion confounding this relation­ship.  The two factors that jump to mind are exercise and weight.  In Framingham, for example, we found that the people who ate the most cholesterol, ate the most saturated fat, ate the most calories, weighed the least, and were the most physi­cally active. This article showed that the people who eat the most nuts weigh the least.  However, in the Cox model, neither exercise nor weight explained the impact on coronary disease.  As to what other factors as­sociated with nut eating explain the benefit, the authors give us a pre­view of a feeding trial, using wal­nuts, that at least shows a favorable change in the blood lipids eating nuts.  Is this due to the polyunsaturated or monounsaturated fat in nuts?  Is it some exotic fiber compo­nent?  Hopefully, this anecdote will allay speculation about some other exotic confounder like television watching, nose-picking, or any myr­iad number of factors not routinely measured in this study.

I suppose that the ARCHIVES will be bombarded by the usual letters about an article such as this by those cheer­ful folk who will want to know if the fall in coronary death rate in the nut eaters was offset by an increase in ac­cidental, violent, or suicidal death.  Per­haps there was a social price to pay, at least from the peanuts that, after all, are legumes.

Should dietitians everywhere trem­ble?  Has the magic bullet arrived?  Is it the humble nut?  Should fat people eat fat-rich nuts to lose weight and athero­sclerosis, or do nuts only work in veg­etarians?  Should nuts replace oat bran as the shield that I can load up on each day which will let those hot dogs just bounce right off my chest, no trouble.

Will this article affect the stock market sending the lipid-drugs reeling when that well-balanced report ap­pears in the Wall Street Journal?  So many questions, so little time.  In the meantime, hold the cheese, I will have my nuts (walnuts?) on whole wheat, please.

william P. castelli, MD.

Framingham Heart Study

National Heart, Lung, and Blood Institute

5 Thurber St

Framingham, MA 01701



1. Fraser GE, Sabato J, Beeson WL, Strahan TM.  A possible protective effect of nut consumption on risk of coronary heart disease; The Adventist Health Study. Arch Intern Med. 1992;152:1416-1424.

2. Hegsted DM, McCandy RB, Myer Ml, Stare FJ.  Quantitative effects of dietary fat on serum cholesterol in man.  Am J Clin Nutr. 1965;17:281-295.

3. Keys A, Anderson JT, Crande F.  Predic­tion of serum cholesterol responses of men lo changes in fats in the diet.  Lancet, 1957; 2:959-966.

4. Keys A.  Coronary heart disease in seven countries. Circulation. 1970;41(suppl 1):1-211.

5. Castelli W.  Cholesterol and lipids in the risk of coronary artery disease: The Framingham Heart Study. Can J Cardiol. 1988;4(suppl A):5A-10A.

6. Stamler J, Shekelle R.  Dietary cholesterol and human coronary heart disease.  Arch Pathol Lab Med. 1988;112:1032-1040