The Diet-Heart Hypothesis: Stuck at the Starting Gate

The diet-heart hypothesis is the idea that (1) dietary saturated fat, and in some versions, dietary cholesterol, raise blood cholesterol in humans and (2) therefore contribute to the risk of heart attack.

I'm not going to spend a lot of time on the theory in relation to dietary cholesterol because there really isn't much evidence to debunk in humans. As far as I can tell, most diet-health researchers don't take this theory seriously anymore because the evidence has simply failed to materialize. Dr. Walter Willett doesn't believe it, and even Dr. Ancel Keys didn't believe it. Here's a graph from the Framingham Heart study (via the book Prevention of Coronary Heart Disease, by Dr. Harumi Okuyama et al.) to drive home the point. Eggs are the most concentrated source of cholesterol in the American diet. In this graph, the "low" group ate 0-2 eggs per week, the "medium" group ate 3-7, and the "high" group ate 7-14 (click for larger image): The distribution of blood cholesterol levels between the three groups was virtually identical. The study also found no association between egg consumption and heart attack risk. Dietary cholesterol does not raise serum cholesterol in the long term, because humans are adapted to eating cholesterol. We simply adjust our own cholesterol metabolism to compensate when the amount in the diet increases, like dogs. Rabbits don't have that feedback mechanism because their natural diet doesn't include cholesterol, so feeding them dietary cholesterol increases blood cholesterol and causes vascular pathology.

The first half of the diet-heart hypothesis states that eating saturated fat raises blood cholesterol. This has been accepted without much challenge by mainstream diet-health authorities for nearly half a century. In 1957, Dr. Ancel Keys proposed a formula (Lancet 2:1959. 1957) to predict changes in total cholesterol based on the amount of saturated and polyunsaturated fat in the diet. This formula, based primarily on short-term trials from the 1950s, stated that saturated fat is the primary dietary influence on blood cholesterol.

According to Keys' interpretation of the trials, saturated fat raised, and to a lesser extent polyunsaturated fat lowered, blood cholesterol. But there were serious flaws in the data from the very beginning, which were pointed out in this searing 1973 literature review in the American Journal of Clinical Nutrition (free full text).

The main problem is that the controlled trials typically compared saturated fats to omega-6 linoleic acid (LA)-rich vegetable oils, and when serum cholesterol was higher in the saturated fat group, this was most often attributed to the saturated fat raising blood cholesterol rather than the LA lowering it. When a diet high in saturated fat was compared to the basal diet without changing LA, often no significant increase in blood cholesterol was observed. Studies claiming to show a cholesterol-raising effect of saturated fat often introduced it after an induction period rich in LA. Thus, the effect may have more to do with LA lowering blood cholesterol than saturated fat raising it. This is not at all what I was expecting to find when I began looking through the short-term trials.

I recently read a 2003 study that addresses this point directly. Muller et al. (free full text) compared the effects of three controlled diets on the blood cholesterol of 25 healthy women. The diets were:

1. High in saturated fat from coconut, low in LA
2. Same as #1, with half the saturated fat replaced by carbohydrate
3. Low in saturated fat, high in LA, with the same total fat as in #1
   

The diets were fed to the whole group for three week periods. Investigators found that diet #3 lowered cholesterol and LDL relative to diets #1 and #2. The total cholesterol of women on diets #1 and #2 were not statistically different (p= 0.09), and their LDL was virtually identical. Thus, a very large difference in saturated fat intake didn't affect total cholesterol or LDL when it was replaced by carbohydrate, but it did when it was replaced by LA. The most straightforward explanation is that LA lowers cholesterol and LDL, but saturated fat has little or no effect on either. From the discussion section:

The most important finding of this study was that lowering total saturated fat in the form of coconut oil, from 22.7 to 10.5 E% without change in the P/S ratio [polyunsaturated to saturated ratio], did not lower total or LDL cholesterol, but significantly reduced HDL cholesterol.

Among the heaps of poorly conducted studies, I was able to find one apparently well-controlled counterexample: Arterioscler. Thromb. Vasc. Biol. 18:441. 1988. In this 8-week study, increasing saturated fat (at the expense of carbohydrate and with LA constant) increased total cholesterol and LDL, while also increasing HDL, and decreasing Lp(a) and triglycerides. Decreasing saturated fat from 15% to 6% of calories (drastic), reduced total cholesterol by 9% and LDL by 11% (calculated by the Friedewald equation). The variation between trials may have to do with the specific saturated fatty acids used in each trial, their duration, or some other unknown confounder.

Reading through the short-term controlled trials, I was struck by the variability and lack of agreement between them. Some of this was probably due to a lack of control over variables and poor study design. But if saturated fat has a dominant effect on serum cholesterol in the short term, it should be readily and repeatably demonstrable. It clearly is not, so I'm left wondering why diet-health authorities are so certain of themselves on this point.

The long-term data are also not kind to the diet-heart hypothesis. Reducing saturated fat while greatly increasing LA does lower blood cholesterol substantially. This was the finding in the well-controlled Minnesota Coronary Survey trial, for example (14% reduction). But in other cases where LA intake changed less, such as MRFIT, the Women's Health Initiative Diet Modification trial and the Lyon Diet-Heart trial, reducing saturated fat intake had little or no effect on total cholesterol or LDL (0-3% reduction). This generally dumbfounded the investigators. The small changes that did occur could easily have been due to other factors, such as increased fiber and phytosterols, since these were multiple-factor interventions.

Another blow to the idea that saturated fat raises cholesterol in the long term comes from observational studies. Here's a graph of data from the Health Professionals Follow-up study, which followed 43,757 health professionals for 6 years (via the book Prevention of Coronary Heart Disease by Dr. Harumi Okuyama et al.): What this graph shows is that at a relatively constant LA intake, neither saturated fat intake nor the ratio of LA to saturated fat were related to blood cholesterol in freely living subjects. This was true across a wide range of saturated fat intakes (7-15%). If we can't even find a consistent association between dietary saturated fat and blood cholesterol in observational studies, how can we claim that saturated fat is a dominant influence on blood cholesterol?

There's more. If saturated fat were important in determining the amount of blood cholesterol in the long term, you'd expect populations who eat the most saturated fat to have high blood cholesterol levels. But that's not at all the case. The Masai traditionally get almost 2/3 of their calories from milk fat, half of which is saturated. In 1964, Dr. George V. Mann published a paper showing that traditional Masai warriors eating nothing but very fatty milk, blood and meat had an average cholesterol of 115 mg/dL in the 20-24 year age group. For comparison, he published values for American men in the same age range: 198 mg/dL (J. Atherosclerosis Res. 4:289. 1964). Apparently, eating three times the saturated animal fat and several times the cholesterol of the average American wasn't enough to elevate their blood cholesterol. What does elevate the cholesterol of a Masai man? Junk food.

Now let's swim over to the island of Tokelau, where the traditional diet includes nearly 50% of calories from saturated fat from coconut. This is the highest saturated fat intake of any population I'm aware of. How's their cholesterol? Men in the age group 20-24 had a concentration of 168 mg/dL in 1976, which was lower than Americans in the same age group despite a four-fold higher saturated fat intake. Tokelauans who migrated to New Zealand, eating half the saturated fat of their island relatives, had a total cholesterol of 191 mg/dL in the same age group and time period, and substantially higher LDL (J. Chron. Dis. 34:45. 1981). Sucrose consumption was 2% on Tokelau and 13% in New Zealand. I think fructose (which makes up 50% of sucrose-- or table sugar-- and 55% of high-fructose corn syrup) is a more logical explanation for the high serum cholesterol and LDL of modern affluent societies, particularly considering the results of this study.

The inevitable conclusion is that if saturated fat influences total cholesterol or LDL concentration at all, the effect is modest and is dwarfed by other factors.

Diet Modification Trials: Notes on Study Design

The other day, my internet meanderings brought me back to a review of fat modification trials conducted by the Cochrane collaboration. This is a not-for-profit group known for its rigorous meta-analyses.

They selected 27 studies that reduced saturated fat or total fat (in some cases along with increased PUFA), and fit several inclusion criteria. The results:

There was no significant effect on total mortality (rate ratio 0.98, 95% CI 0.86 to 1.12), a trend towards protection form cardiovascular mortality (rate ratio 0.91, 95% CI 0.77 to 1.07), and significant protection from cardiovascular events (rate ratio 0.84, 95% CI 0.72 to 0.99). The latter became non-significant on sensitivity analysis.

Trials where participants were involved for more than 2 years showed significant reductions in the rate of cardiovascular events and a suggestion of protection from total mortality ["suggestion" = not statistically significant]. The degree of protection from cardiovascular events appeared similar in high and low risk groups, but was statistically significant only in the former.

"Sensitivity analysis" is a statistical method that allows investigators to take into account the limitations of their model, in this case, the way in which they aggregated the studies' data. Another way of putting their findings is that they found no significant effect of fat modification on mortality or cardiovascular mortality, and they found a reduction in non-fatal "cardiovascular events" (more on this phrase later) only in a subset of the data.

I'll be the first to admit the meta-analysis isn't perfect. They cast too wide a net, not allowing them to distinguish the effect of reducing total fat from the effect of reducing saturated fat. They lumped both together, which from a practical standpoint isn't actually a problem because both sets of studies show essentially the same thing: zilch. But it's still not the best way to conduct a meta-analysis. They also omitted the Sydney Diet-Heart study for mysterious reasons, which was a five year randomized trial that found an increase in mortality in volunteers substituting vegetable oils for animal fat. Then there's the conclusion, which boggles the mind:

Lifestyle advice to all those at high risk of cardiovascular disease (especially where statins are unavailable or rationed), and to lower risk population groups, should continue to include permanent reduction of dietary saturated fat and partial replacement by unsaturates.

Are these the same people who wrote the results section? I don't understand how they arrived at that conclusion from their own results.

In any case, this brings me to my main point. There are two types of outcomes you can measure in these trials: "hard endpoints" and "soft endpoints". Hard endpoints are not subjective. The hardest endpoint is death. Either you're dead or you aren't; there's no room for interpretation there. A bit less hard is death from a particular cause, such as heart attack. In that case, you're definitely dead, but the physician has to guess what you died of. That involves some judgment on the part of the physician and can introduce bias, depending on the study design. The softest endpoints are non-fatal events like angina, bypasses and stents. These depend on the judgment of both the physician and the patient, and are the most susceptible to bias.

The gold standard for controlled trials is a design known as "double-blind", where neither the participant nor the physician knows which group the participant is in. This design eliminates bias from both the participant and the physician side, allowing correction for the placebo effect and subtle bias in diagnosis. This is easy to do for drug trials, where placebo pills look just like the drug. But it's more difficult to pull off in a diet trial, where the patient knows what foods he's eating. Still, it can be done by giving participants similar-looking margarines containing either saturated or polyunsaturated fats, or sometimes by controlling diets in an institutionalized setting.

There have been three double-blind trials comparing the incidence of heart attack and/or mortality in volunteers eating either saturated animal fat or polyunsaturated vegetable fat: the 1968 National Diet-Heart trial (2 years), the 1969 Los Angeles Veterans' Administration trial (8 years), and the 1989 Minnesota Coronary Survey trial (4.5 years). The two studies that reported total mortality found no significant difference between groups. Two out of three found no difference in heart attack deaths. Of the two that reported on non-fatal cardiovascular endpoints, one found a significant difference. The V.A. trial was the only one to find a significant difference in heart attack deaths (18% decrease) and non-fatal events. There were significantly more heavy smokers in the animal fat arm of the V.A. trial, which was an unfortunate consequence of the participant randomization process. So that result is difficult to interpret.

The three double-blind diet trials, with the least potential for bias, really give no support to the idea that saturated/animal fat contributes to cardiovascular disease. As the participants were already eating a diet high in omega-6 to begin with, there is also no detectable effect of increasing omega-6 on cardiovascular health.

Many of the trials of this nature have been "single-blinded", where the participant knows which group he's in, but the physician doesn't. In this case, all endpoints involving death will be unbiased, because the physician deciding the diagnosis is not influenced by knowing what group the patient is in. However, soft, non-fatal events such as angina are still highly susceptible to the placebo effect. This is because they depend on the subjective judgment of the patient, who knows which group he's in.

I think it's interesting to note that very few dietary fat modification trials have found reductions in total mortality, which is the hardest endpoint and the least susceptible to bias. This is reflected in the Cochrane collaboration's findings. However, a number of the non-blinded and single-blinded studies have found differences in non-fatal cardiovascular events, sometimes creating absurd results. For example, in the 1966 Anti-Coronary Club trial, participants in the vegetable oil group had a significant reduction in non-fatal cardiovascular events, but a massive increase in cardiovascular deaths and total mortality. The former result could result from a placebo effect, due to the non-blinded nature of the trial.

The fact that the Cochrane review found statistically significant benefits of fat modification in soft, non-fatal endpoints (for a portion of the data set), but not endpoints involving death, suggests to me that what we're seeing may actually be a placebo effect resulting from the fact that patients were not blinded in the majority of these trials.

The only "fat modification" intervention that consistently reduces total and cardiovascular mortality is omega-3 fat supplementation, ideally in combination with omega-6 restriction. This is supported by the results of the DART trial, the Lyon Diet-Heart trial, the ISIS trial and the the GISSI-prevenzione trial.