ApoB

What it actually measures

Picture a fleet of delivery vans moving cholesterol around your body. Now picture a road where those vans occasionally get stuck in the wall and start a slow pile-up that becomes plaque.

Here is the question that matters. Is the risk about how much cargo is being shipped, or about how many vans are on the road?

A standard lipid panel answers the first question. LDL-C measures the cholesterol carried inside your LDL particles, added up. It is essentially the weight of the cargo.

ApoB answers the second. Every one of those atherogenic particles carries exactly one apoB molecule on its surface. One particle, one apoB, no exceptions. So an apoB measurement is a direct count of the vans.

And it turns out the vans are what matters, because a particle enters an artery wall as a particle, regardless of how full it happens to be.

That is the whole idea. Not a competing theory of cholesterol, not an alternative model. The same particles, counted rather than weighed.

Normal and optimal are different questions

Your lab will print something like 40 to 125 mg/dL next to your apoB result, and that range is close to useless for the purpose you brought to it.

It is a population range. It describes the middle of the people who got tested, and since cardiovascular disease is the leading cause of death across most of the developed world, describing the middle of that population is not describing a healthy target.

The numbers that actually get used come from lipid guidelines instead, and they are risk-stratified rather than universal. Commonly cited: under about 90 mg/dL at average risk, under roughly 80 at higher risk, and lower again in established disease.

I want to be careful with those figures because they are not one agreed set. Different guideline bodies land in different places, and the right target for one person genuinely depends on their overall risk rather than on a single number.

So the useful framing is not a magic apoB. It is that the printed reference range and the risk target are answering two completely different questions, and the gap between them is large enough to matter.

When the two numbers disagree

Here is the fact that should get your attention.

Your LDL cholesterol and your apoB usually agree. Usually. When they do, either one tells you roughly what you need to know and the distinction is academic.

The interesting question is what happens when they disagree, which researchers call discordance, and there is a clever reason to study exactly that. LDL-C and apoB are so tightly correlated that ordinary statistics struggle to separate them. Discordance analysis gets round it by deliberately isolating the people in whom the two markers point different directions and asking which one was right.

A 2025 systematic review in the Journal of Clinical Lipidology gathered every such study it could find. Fifteen studies, 593,354 participants, populations on and off statin therapy, using four different analytical approaches.

ApoB outperformed LDL-C in 9 of 9 comparisons. Against non-HDL cholesterol, apoB came out significantly more accurate in 7 studies. The authors concluded that neither LDL-C nor non-HDL-C is an adequate clinical surrogate for apoB.

Now the honest complication, because I would rather you had the whole picture. A 2026 Copenhagen General Population Study analysis in JAMA Cardiology, following 94,398 people for a median of over thirteen years, found non-HDL cholesterol and apoB associated with similar risk on continuous scales. So the specific question of apoB versus non-HDL cholesterol is still live in the literature.

What is not really in dispute is the comparison people actually care about, which is apoB against the LDL-C on their existing lab report. There the evidence is consistent and it points one way.

And here is the practical part. Discordance is not random. It concentrates in people with insulin resistance, high triglycerides, metabolic syndrome and type 2 diabetes, because those conditions produce more particles carrying less cholesterol each. Which means the reassuring LDL-C is least trustworthy in exactly the group most likely to be told their cholesterol is fine.

What moves it

Down: reducing saturated fat intake, which is the most direct dietary lever on these particles. Soluble fibre, from oats, legumes, psyllium and fruit, which pulls bile acids out and makes the liver clear more particles. Losing excess visceral fat. Improving insulin sensitivity through exercise and diet, which matters especially here because insulin resistance raises particle count out of proportion to cholesterol content. Treating hypothyroidism where present. And lipid-lowering medication, which is a prescribing decision.

Up: genetics, including familial hypercholesterolaemia, which is far more common than most people realise and is worth knowing about. Insulin resistance and high triglycerides. Excess visceral fat. Untreated hypothyroidism. Kidney disease. And a diet pattern high in saturated fat.

The honest hedge

I am not a doctor and I am not diagnosing anyone, and cardiovascular risk is one of the areas where that matters most.

ApoB is a measurement rather than a verdict. It tells you about one dimension of risk, and the whole picture includes blood pressure, blood sugar, smoking, family history, and whether there is already plaque present, which a calcium score or an imaging study answers and a blood test cannot.

I also want to be plain about what I am not doing here. Whether to take lipid-lowering medication is a genuine clinical decision with real trade-offs, and it belongs with a physician who knows your full risk picture. Nothing on this page is an argument for or against it.

What I will say confidently is that measuring apoB gives you a more accurate count than inferring it from LDL-C, that the discordance research is large and consistent on that point, and that the people most likely to be misled by a comfortable LDL-C are the ones with insulin resistance. That is worth one extra line on a lab order.

So. What to actually do.

Tonight, at no cost. Pull up your last lipid panel. If it shows total cholesterol, LDL-C, HDL-C and triglycerides and nothing else, your particle count has never been measured.

Check your triglyceride to HDL ratio while you are there. You can calculate it from numbers you already have, and a high ratio is the flag that says your LDL-C is more likely to be discordant.

When you test, order the advanced panel. An advanced lipid panel includes apoB alongside the standard numbers, so you get the count and the cargo weight from one draw.

Pair it with a metabolic marker. Fasting insulin or HbA1c, because insulin resistance is what drives the discordance in the first place and it is the more addressable of the two problems.

Work the food levers that actually move it. Soluble fibre daily, saturated fat down, and enough movement to shift insulin sensitivity. Those are free and they move this number.

Take the whole picture to a doctor. ApoB is one input into a risk conversation that also includes blood pressure, family history and possibly imaging. It is not a decision on its own.

None of this has to happen this week. But tonight you can work out your triglyceride to HDL ratio from a report you already have.

You are not a cholesterol number. You are a circulation, moving a fleet of particles through a system that has been counting them all along, whether or not anyone handed you the count.

Your body is not broken. It is blocked. And in this case that is not a metaphor, it is a slow literal accumulation, driven by a quantity most lipid panels never report.

Go count the trucks.