ONE WILD WORLD — NEWSLETTER

ISSUE 053 · DEEP DIVE · MYTHS & MISINFORMATION

Of everything humans eat, exactly one item has ever been named and restricted by public health authorities specifically to bring down heart disease. Not bacon. Not butter. The egg.

In 1968 the American Heart Association told people to eat no more than three yolks a week. A cap of 300 milligrams of cholesterol a day followed and stayed in the official American guidelines for decades. Generations of people ordered the omelette without the yolk and felt virtuous about it.

Then in 2015 the advisory committee behind those guidelines concluded that cholesterol is not a nutrient of concern for overconsumption, and the 300 mg cap was dropped. No numeric limit remains. In 2019 the AHA reaffirmed that no useful number exists.

Forty-seven years, one food, and the advice was withdrawn.

The reason it collapsed is almost embarrassingly simple, and it was knowable the whole time.

Most of the cholesterol in your bloodstream was not eaten. Your liver made it. And the liver runs a thermostat: eat more cholesterol, and it produces less to compensate. In roughly 75 to 85 per cent of people that feedback loop works well enough that dietary intake barely moves the needle.

The effect is not literally zero. Pooled studies suggest an extra 100 milligrams of dietary cholesterol a day raises LDL by somewhere around 2 to 5 mg/dL. One egg carries about 186 milligrams. So an egg a day might lift your LDL by four to nine points — real, measurable, and nowhere near the size of the thing that got built on top of it.

Ancel Keys, the man whose name is attached to the whole diet-heart era, worked this out himself: dietary cholesterol had a small effect compared with saturated fat. The advice outlived the evidence for it by decades.

One piece of officialdom still has not caught up. The nutrition label on American packaging continues to list cholesterol against a daily value of less than 300 mg — the number the guidelines abandoned.

Now the second mistake, and this one is more interesting because it fooled the professionals rather than the public.

HDL got the nickname “good cholesterol” because people with more of it have less heart disease. That observation is solid. The inference drawn from it — raise HDL, get less heart disease — is where it went wrong.

In 2012, Voight and colleagues published a study in the Lancet using the same genetic method that had confirmed LDL. Variants that raise LDL increased the risk of heart attack, as expected. Variants that raise HDL showed no relationship with heart disease risk at all.

The drugs told the same story, more expensively. Torcetrapib, the first CETP inhibitor into clinical development, pushed HDL up substantially. It did not reduce cardiovascular risk. It increased it. At least three drugs in that class have failed to demonstrate protection. Niacin did the same thing in the AIM-HIGH trial: added to a statin, it raised HDL and lowered triglycerides, and cut precisely nothing off the event rate.

Here is the caveat, because the tidy version of this story is also wrong. More recent work suggests those CETP failures were probably about the specific compounds rather than the target itself, and drug-target genetic analysis hints that properly inhibiting CETP might yet help. So the honest statement is narrower than the internet version.

Which brings us to LDL in the blood, where the mess ends and the evidence starts.

Three independent lines of evidence converge here, and they have different weaknesses, which is what makes the convergence matter.

Population studies show the association. On their own they prove nothing, because people with high LDL differ from people with low LDL in a hundred other ways.

Mendelian randomisation fixes that. The trick is that your genetic variants are dealt out essentially at random when you are conceived, long before diet, income, smoking or exercise get involved. So comparing people who inherited slightly higher LDL with people who inherited slightly lower LDL is close to a randomised trial that nature ran for you, lasting your entire life. These studies consistently find LDL causally linked to coronary heart disease. One large analysis put the odds ratio at 1.45 per standard deviation of genetically higher LDL.

And randomised drug trials close it. Lower LDL pharmacologically and heart attacks fall. That is not correlation. That is intervention.

The European Atherosclerosis Society consensus adds the detail that makes it click: the effect is both causal and cumulative. The longer the exposure runs, the steeper the risk curve gets. It is not your LDL this year. It is your LDL multiplied by all the years you have had it.

Worth explaining, since it is the most prescribed drug on the planet and most people have no idea of the mechanism.

Your liver builds cholesterol through a chain of reactions, and one enzyme near the start — HMG-CoA reductase — sets the pace. A statin blocks it. The liver, now short of supply, responds by putting more LDL receptors on its own cell surfaces, and those receptors pull LDL out of your bloodstream. Blood LDL typically drops by 30 to 50 per cent.

Atorvastatin, simvastatin, rosuvastatin. Same trick, different molecules.

So why does the internet think cholesterol is a hoax?

Because of 2015. The headlines that year said the cholesterol guidelines had been wrong, which was true, and almost none of them made clear that this applied to cholesterol in food and not cholesterol in blood. The correction was real. The generalisation drawn from it was not.

The deeper error is older than that, and it is baked into the words. We call all of it “cholesterol”, as though it were a single substance you could have too much or too little of. It is not. LDL and HDL are not two types of cholesterol — they are particles that carry it around, and what matters is the traffic, not the cargo. Give one particle a nickname like “bad” and another “good” and you have built a folk theory that will take fifty years to dismantle.

Ference, B. A. et al. (2017). “Low-density lipoproteins cause atherosclerotic cardiovascular disease. 1.” Consensus statement, European Atherosclerosis Society Consensus Panel. European Heart Journal 38(32), 2459–2472.

Voight, B. F. et al. (2012). “Plasma HDL cholesterol and risk of myocardial infarction: a mendelian randomisation study.” The Lancet 380, 572–580.

Keene, D. et al. (2014). “Effect on cardiovascular risk of high density lipoprotein targeted drug treatments niacin, fibrates, and CETP inhibitors.” BMJ 349, g4379.

Schmidt, A. F. et al. (2021). “Cholesteryl ester transfer protein (CETP) as a drug target for cardiovascular disease.” Nature Communications 12.

Scientific Report of the 2015 Dietary Guidelines Advisory Committee, US Department of Health and Human Services / USDA (2015).

Carson, J. A. S. et al. (2020). “Dietary Cholesterol and Cardiovascular Risk.” AHA Science Advisory. Circulation 141(3).

Griffin, J. D. & Lichtenstein, A. H. (2013). “Dietary Cholesterol and Plasma Lipoprotein Profiles.” Current Nutrition Reports 2, 274–282.

AIM-HIGH Investigators (2011). “Niacin in patients with low HDL cholesterol levels receiving intensive statin therapy.” New England Journal of Medicine 365, 2255–2267.

ONE MORE THING

You can read our MANY newsletters for free at our website:

Read our archive for free at OneWildWorld.com →