GCBC is pretty contrarian in it’s health advice. It doesn’t come out and say you should change your diet in any specific way, but it does call into question a lot of the things I had always assumed were indisputable.

Salt

The most surprising part of GCBC for me was the very weak link between salt and high blood pressure. The medical wisdom seems to be that if you eat salt, your body will have to hang on to more water in order to keep your blood at the right pH. The research cited in GCBC seems to show that this does happen, but not actually all that much.

The best you can hope to do by halving your salt intake is about 2-5mmHg lowering in blood pressure. Given that the safe range of blood pressure is 100-140mmHg systolic and 60-90mmHg diastolic, a change of 5mmHg doesn’t seem that significant. Salt probably won’t push you out of the healthy range, so there must be something else going on.

The alternative proposed in GCBC is that eating lots of carbohydrates is actually what causes hypertension. The idea is that eating carbohydrates causes insulin spikes, which causes the kidneys to retain salt (regardless of salt in diet), which causes the body to retain water, which causes high blood pressure.

This is a pretty long chain of cause and effect, but apparently cutting out carbohydrates has been shown to work as well as using diuretics to reduce blood pressure. Hypertension is also often associated with insulin-resistance, which implies some kind of effect from carbohydrates.

Fat

The other controversial suggestion in GCBC is that eating fatty foods doesn’t cause or contribute to heart disease. In fact, a huge chunk of the book is spent in support of this assertion. You should read the whole book if you want all the details, but the main idea seems to hinge on a confusion of what cholesterol is and how it works in the body.

The historical perspective has been that eating fat leads directly an increase in blood cholesterol. Blood cholesterol then builds up in arteries, forming plaques. These plaques create bottlenecks in arteries that can then be blocked, leading to heart attacks. That story is the main impetus for the huge push to lower fat intake over the past 50 years.

According to GCBC, the science of cholesterol has progressed significantly from where it was in the mid-1950s.

It turns out that cholesterol is a critical ingredient in cell membranes, some hormones, and other functions. Our bodies will make it when we need it, but if we get it in our diets then our bodies will just use that instead. It gets made in the liver or taken in from food and transported by blood to the various tissues of the body. However, cholesterol doesn’t dissolve well in water. It turns out it needs to be transported in the blood by lipoprotiens.

Each lipoprotien serves as a carrier for a bunch of cholesterol (or some other fatty molecule). Lipoprotiens come in different sizes and types, smaller ones holding less cholesterol. The less protien you have, the smaller your lipoprotiens. This is what the so called HDLs, LDLs and VLDLs are. They’re not cholesterol at all, they’re the cholesterol transport mechanisms.

HDLs transfer fatty molecules away from cells that don’t want them by interacting with a binding site on the outside of the cell. HDLs are not associated with having heart disease.
LDLs transfer fatty molecules to the cells that want them. Cells that want fatty molecules will create receptacles on their membrane that will accept LDL molecules. LDLs are associated with having heart disease.
IDLs apparently get turned into large LDLs, which aren’t that likely to give you heart disease.
VLDLs apparently get metabolized into small LDLs. VLDLs themselves are associated with heart disease. The small LDLs that are created from VLDL are also associated with heart disease.

I think GCBC didn’t quite get this right. You need both HDL and LDL because they do different jobs. What GCBC does seem to be right about is that LDL (and thus IDL and VLDL) itself is associated with heart disease, but cholesterol doesn’t seem to be. I don’t think there’s much scientific dispute about this now, but I do think that medical researchers are trying to cover their asses for all that talk about cholesterol 50 years ago. Why else would the wikipedia cholesterol page spend 10 long paragraphs talking about LDL and HDL with only a small note at the beginning stating that they aren’t actually cholesterol?

So if fat affects cholesterol (which it does seem to), then what does it do to the amount of IDL and VLDL? It seems like what we want is probably IDL instead of VLDL to protect ourselves from heart disease. The claim in GCBC is that eating carbohydrates leads to lots of VLDL and not much IDL. Eating fat will have a small effect on cholesterol, but no effect at all on IDL and VLDL levels.

So the takeaway according to GCBC is don’t worry too much about fat in the diet.