Eicosanoids and Ischemic Heart Disease, Part II

Here's where it gets more complicated and more interesting. The ratio of omega-6 to omega-3 matters, but so does the total amount of each. This is a graph from a 1992 paper by Dr. Lands:

Allow me to explain. These lines are based on values predicted by a formula developed by Dr. Lands that determines the proportion of omega-6 in tissue HUFA (highly unsaturated fatty acids; includes 20- to 22-carbon omega-6 and omega-3 fats), based on dietary intake of omega-6 and omega-3 fats. This formula seems to be quite accurate, and has been validated both in rodents and humans. As a tissue's arachidonic acid content increases, its EPA and DHA content decreases proportionally.

On the Y-axis (vertical), we have the proportion of omega-6 HUFA in tissue. On the X-axis (horizontal), we have the proportion of omega-6 in the diet as a percentage of energy. Each line represents the relationship between dietary omega-6 and tissue HUFA at a given level of dietary omega-3.

Let's start at the top. The first line is the predicted proportion of omega-6 HUFA in the tissue of a person eating virtually no omega-3. You can see that it maxes out around 4% of calories from omega-6, but it can actually be fairly low if omega-6 is kept very low. The next line down is what happens when your omega-3 intake is 0.1% of calories. You can see that the proportion of omega-6 HUFA is lower than the curve above it at all omega-6 intakes, but it still maxes out around 4% omega-6. As omega-3 intake increases, the proportion of omega-6 HUFA decreases at all levels of dietary omega-6 because it has to compete with omega-3 HUFA for space in the membrane.

In the U.S., we get a small proportion of our calories from omega-3. The horizontal line marks our average tissue HUFA composition, which is about 75% omega-6. We get more than 7% of our calories from omega-6. This means our tissue contains nearly the maximum proportion of omega-6 HUFA, creating a potently inflammatory and thrombotic environment!
This is a very significant fact, because it explains three major observations:
  1. The U.S has a very high rate of heart attack mortality.
  2. Recent diet trials in which saturated fat was replaced with omega-6-rich vegetable oils didn't cause an increase in mortality, although some of the very first trials in the 1960s did.
  3. Diet trials that increased omega-3 decreased mortality.
Observation number two is used by proponents of PUFA-rich vegetable oils, and it's a fair point. If omega-6 causes heart attacks, why hasn't that shown up in controlled trials? Here's the rebuttal. First of all, it did show up in two of the first controlled trials in the 1960s: Rose et al., and the unfortunately-named Anti-Coronary Club trial. In the first, replacing animal fat with corn oil caused a 4-fold increase in heart attack deaths and total mortality. In the second, replacing animal fat with polyunsaturated vegetable oil increased heart attack death rate, and total mortality more than doubled.

But the trend didn't continue into later trials. This makes perfect sense in light of the rising omega-6 intake over the course of the 20th century in the U.S. and other affluent nations. Once our omega-6 intake crossed the 4% threshold, more omega-6 had very little effect on the proportion of omega-6 HUFA in tissue. This may be why some of the very first PUFA diet trials caused increased mortality: there was a proportion of the population that was still getting less than 4% omega-6 in its regular diet at that time. By the 1980s, virtually everyone in the U.S. (and many other affluent nations) was eating more than 4% omega-6, and thus adding more did not significantly affect tissue HUFA or heart attack mortality.

If omega-3 intake is low, whether omega-6 intake is 5% or 10% doesn't matter much for heart disease. At that point, the only way to reduce tissue HUFA without cutting back on omega-6 consumption is to outcompete it with additional omega-3. That's what the Japanese do, and it's also what happened in several clinical trials including the DART trial.

This neatly explains why the French, Japanese and
Kitavans have low rates of ischemic heart disease, despite the prevalence of smoking cigarettes in all three cultures. The French diet traditionally focuses on animal fats, eschews industrial vegetable oils, and includes seafood. They eat less omega-6 and more omega-3 than Americans. They have the lowest heart attack mortality rate of any affluent Western nation. The Japanese are known for their high intake of seafood. They also eat less omega-6 than Americans. They have the lowest heart attack death rate of any affluent nation. The traditional Kitavan diet contains very little omega-6 (probably less than 1% of calories), and a significant amount of omega-3 from seafood (about one teaspoon of fish fat per day). They have an undetectable incidence of heart attack and stroke.

In sum, this suggests that an effective way to avoid a heart attack is to reduce omega-6 consumption and ensure an adequate source of omega-3. The lower the omega-6, the less the omega-3 matters. This is a nice theory, but where's the direct evidence? In the next post, I'll discuss the controlled trial that proved this concept once and for all: the Lyon diet-heart trial.