Apolipoprotein B (apoB) counts the atherogenic particles that actually drive atherosclerosis — a more direct measure of risk than the cholesterol those particles happen to carry. This piece sets out how to interpret an apoB result, where it diverges from LDL-C, and how to lower it.

For most of the last three decades, cardiovascular risk has been read through a single number: LDL cholesterol.

It has served us reasonably well. It is cheap, familiar, embedded in every lab report and every guideline, and it responds predictably to treatment. But LDL-C measures the cholesterol carried inside low-density particles — not the number of particles themselves.

And it is the particles that cause the disease.

Every atherogenic lipoprotein — LDL, VLDL and its remnants, IDL, and Lp(a) — carries exactly one molecule of apolipoprotein B. So a single apoB measurement counts the total number of particles capable of entering and lodging in the arterial wall. That is a more direct reading of the thing we are actually trying to prevent.

The distinction sounds academic. In practice, it changes how confidently we can assess and treat an individual patient.

What apoB actually measures

apoB-100 is the structural protein on every particle produced by the liver that contributes to atherosclerosis. One particle, one apoB. Because the count is stoichiometric, plasma apoB is effectively a direct measure of atherogenic particle number, dominated in most people by LDL.

This matters because cholesterol content per particle is not fixed. Two patients can carry the same quantity of LDL cholesterol in very different numbers of particles. The one packaging that cholesterol into more numerous, smaller, denser particles has more apoB — and a higher risk — despite an identical LDL-C on paper.

LDL-C tells you how much cholesterol is travelling. apoB tells you how many vehicles are carrying it. The vehicle count is what tracks with events.

Where LDL-C and apoB diverge — and why it matters

For most patients, apoB, LDL-C and non-HDL-C move together, and any of them will do. The problem is the minority in whom they don't — and that group is neither small nor easy to identify in advance.

Large discordance analyses are now consistent on this point. Across a wide range of LDL-C values, apoB varies substantially between individuals; the same LDL-C can correspond to meaningfully different apoB levels and therefore meaningfully different 10-year ASCVD risk. A pooled discordance review found apoB outperformed LDL-C as a risk marker in 9 of 9 studies, and was superior to non-HDL-C in the majority — the conclusion being that neither LDL-C nor non-HDL-C is an adequate surrogate for apoB at the individual level.

The groups most prone to discordance are exactly the ones we see most often:

Insulin resistance, metabolic syndrome and type 2 diabetes — cholesterol-depleted, particle-rich profiles where LDL-C understates particle burden.
Hypertriglyceridaemia — more cholesterol shifts into triglyceride-rich remnants, so LDL-C drifts away from true particle count.
Low or "normal" LDL-C on treatment — a reassuring LDL-C can sit alongside a still-elevated apoB, flagging residual risk that the standard panel misses.

The clinical consequence is simple: when LDL-C is low but apoB (or non-HDL-C) is disproportionately high, the patient is carrying more small, dense, atherogenic particles than the LDL-C implies — and is being under-treated if we stop at the cholesterol number.

Interpreting an apoB result

apoB is most useful in two situations: refining risk when the standard panel is ambiguous or discordant, and confirming the adequacy of treatment once therapy is established. It is particularly worth measuring in patients with diabetes, metabolic syndrome, raised triglycerides, or a low LDL-C that doesn't fit the overall clinical picture.

Targets are tied to overall cardiovascular risk rather than read against a single population "normal" range — and this is a common source of error, because laboratory reference ranges are population-derived, not risk-optimised. A result reported as "normal" can still be too high for a given patient.

Broadly aligned with European (ESC/EAS) and National Lipid Association framing:

Risk category	Approximate apoB goal	Corresponding LDL-C goal
Moderate risk	< 1.00 g/L (100 mg/dL)	< 2.6 mmol/L
High risk	< 0.80 g/L (80 mg/dL)	< 1.8 mmol/L
Very high risk (established ASCVD, diabetes with target-organ damage)	< 0.65 g/L (65 mg/dL)	< 1.4 mmol/L

Exact thresholds vary between guidelines and are being revised as apoB's standing rises — recent guidance has moved it from a niche test toward a recommended option for risk refinement, rather than a replacement for LDL-C in every patient. Treat the figures above as working targets and confirm against the current guideline you are practising under.

A few practical points:

apoB does not require a fasting sample, which makes it operationally easier than a full fasting lipid profile.
Recheck roughly 8–12 weeks after any change in therapy.
Where triglycerides remain above ~2.3 mmol/L after statin optimisation, apoB and non-HDL-C are better guides to residual risk than LDL-C.
apoB does not capture Lp(a) mass separately; a markedly elevated apoB with otherwise modest LDL-C should prompt a one-off Lp(a) measurement.

Reduction strategies

Lowering apoB means lowering atherogenic particle number. Every effective LDL-lowering intervention also lowers apoB, though not always to the same degree — which is why apoB is useful for confirming that treatment has done what the LDL-C suggests.

Lifestyle and diet (the foundation, and often underused):

Mediterranean-style pattern; reduce saturated fat to < 7% of energy.
Increase soluble fibre — oats, psyllium, legumes.
Plant sterols/stanols ~2 g/day (~10% LDL-C reduction).
Weight loss, physical activity, and addressing insulin resistance, which directly improves the particle-rich profiles described above.

Pharmacological therapy, layered to target:

Statins remain the foundation — high-intensity regimens lower LDL-C by ~50% and apoB substantially, with the strongest outcome evidence and best cost-effectiveness of any option.
Ezetimibe adds a further ~15–20% LDL-C reduction; a sensible, inexpensive second step.
Bempedoic acid for statin-intolerant patients or as add-on.
PCSK9 monoclonal antibodies (evolocumab, alirocumab) reduce apoB by roughly 47–54% on top of statins; in FOURIER, evolocumab cut major cardiovascular events by ~15%.
Inclisiran (siRNA) offers similar LDL-C/apoB lowering with twice-yearly dosing, which removes the adherence problem that quietly undermines so much lipid therapy.
Fibrates have a modest effect on apoB, mainly relevant in combined hyperlipidaemia.

The principle running through all of this is exposure over time: cumulative particle burden, not a single reading, is what damages the artery. Earlier and sustained lowering matters more than the precise agent.

Where this works — and where it fails — in practice

The evidence that apoB is the more accurate marker has been settled for some time. That is not really the obstacle.

The obstacle is that LDL-C is woven into every layer of the system: the lab panel, the GP recall, the QOF target, the patient's mental model of their own "cholesterol number". apoB is a single, cheap, non-fasting test that resolves a genuine clinical ambiguity — and it still isn't ordered routinely, because nothing in the pathway prompts it and the result doesn't slot neatly into the targets clinicians are measured against.

So the gap here is not knowledge. It is implementation. The test that would most improve risk stratification in the patients who are hardest to assess is the one our workflows are least set up to deliver.

Until ordering apoB becomes the default in the discordant-risk patient — diabetic, metabolic, hypertriglyceridaemic, or simply not adding up — we will keep treating a proxy when the real measure is one box on the request form away.

Key references

National Lipid Association. Role of apolipoprotein B in the Clinical Management of Cardiovascular Risk in Adults: An Expert Clinical Consensus (J Clin Lipidol, 2024).
Marston NA et al. Discordance among apoB, non-HDL-C, and triglycerides: implications for cardiovascular prevention (Eur Heart J, 2024).
Individual Variation in the Distribution of Apolipoprotein B Levels Across the Spectrum of LDL-C or Non-HDL-C (JAMA Cardiology, 2024).
Sniderman AD et al. ApoB, LDL-C, and non-HDL-C as markers of cardiovascular risk — discordance meta-analysis (J Clin Lipidol, 2025).
Sabatine MS et al. Evolocumab and Clinical Outcomes in Patients with Cardiovascular Disease (FOURIER; NEJM, 2017).
Apolipoprotein B: Bridging the Gap Between Evidence and Clinical Practice (Circulation, 2024).

Apo B and Cardiovascular Risk: Clinical Interpretation and Reduction Strategies