ApoB vs LDL-C: Which Cholesterol Number Matters More

Across 233,455 adults in 12 prospective reports, apolipoprotein B beat LDL cholesterol as a marker of cardiovascular risk. One standard deviation higher carried 43% more risk. The same step in LDL-C carried 25%.

Non-HDL cholesterol sat between them, at 34%. That ranking has held for fifteen years.

Your NHS panel gives you LDL-C and non-HDL-C. ApoB you have to ask for, and usually pay for. Here is what the difference is worth.

The two numbers measure different things

LDL-C weighs cargo. ApoB counts vehicles.

Every atherogenic lipoprotein carries exactly one apolipoprotein B molecule. That includes LDL, VLDL, IDL, their remnants and lipoprotein(a). Count apoB and you have counted the particles.

LDL-C instead reports the cholesterol those particles happen to be carrying. In most people the two track closely. In UK Biobank the correlation between apoB and LDL-C is 0.96.

Ninety-six percent agreement sounds like a settled argument. It is not, because the disagreement is not random.

Under insulin resistance, raised triglycerides and central weight gain, particles get smaller and more numerous. Cholesterol per particle drops. LDL-C stays flat while the particle count climbs.

And it is the particle that crosses the endothelium and lodges in the artery wall, not the cholesterol riding inside it.

What the genetics say

Observational data can rank markers. It cannot tell you which one is doing the damage. Mendelian randomisation gets closer, because you inherit your genotype at conception and it is not confounded by how you live.

Richardson and colleagues ran a genome-wide scan of lipid traits in roughly 440,000 UK Biobank adults, then tested them against coronary outcomes from CARDIoGRAMplusC4D. Assessed one at a time, LDL-C, triglycerides and apoB all looked causal.

Put all three in the same model and only one survived.

Ference and colleagues approached it from a different direction, using 654,783 people and 91,129 coronary cases. They compared triglyceride-lowering variants in the LPL gene with LDL-C-lowering variants in the LDLR gene.

Per 10 mg/dL lower apoB, the two scores gave near-identical results. The LPL score returned an odds ratio of 0.771, the LDLR score 0.773.

Two entirely different lipid changes. The same benefit per particle removed. That is a strong hint about what the artery is responding to.

Same LDL-C, very different apoB

This is where a correlation of 0.96 stops being reassuring.

Sniderman and colleagues followed 293,876 UK Biobank adults with no cardiovascular disease for a median of 11 years, recording 19,982 first events. Then they asked a plain question. If you know someone's LDL-C, how well do you know their apoB?

At an LDL-C of 3.4 mmol/L, capturing 95% of people took an apoB range of about 0.86 to 1.09 g/L. That is a spread of roughly a quarter, hidden inside one LDL-C value.

The spread had consequences. In the adjusted model, residual apoB stayed statistically significant after accounting for LDL-C and HDL-C. The residuals of LDL-C, non-HDL-C and triglycerides did not, once apoB was in the model.

The authors' conclusion was blunt: LDL-C, non-HDL-C and triglycerides are not adequate proxies for apoB.

Who LDL-C is most likely to mislead

Discordance is not spread evenly. It clusters, which makes it predictable.

Marston and colleagues followed 389,529 UK Biobank adults who were not on lipid-lowering therapy, for a median 11.1 years. Each standard deviation higher apoB carried 38% more myocardial infarction risk (HR 1.38, 95% CI 1.34-1.42). Adjust fully for triglycerides, non-HDL-C and HDL-C and it still held at 1.27.

Triglycerides behaved differently. Alone, they carried 16% more risk per standard deviation. After adjustment for apoB, that association disappeared.

The same analysis tested whether the kind of particle mattered. It did not. The ratio of triglyceride to LDL cholesterol showed no association with infarction, meaning a triglyceride-rich VLDL particle looks about as atherogenic as an LDL particle.

That is a useful simplification. You do not need to know which particles you are carrying. You need to know how many.

How often the two numbers disagree depends entirely on where you draw the line. The Polish NATPOL survey put the discordance rate anywhere from 0.2% to 45.2%, depending on the thresholds used. That range is not a flaw in the data. It is a warning against treating any single cut-point as a verdict.

That survey of around 2,000 adults found the familiar shape. People with high apoB but unremarkable LDL-C had the features of metabolic syndrome — central weight, raised triglycerides, higher fasting glucose.

So the people most likely to be wrongly reassured by a clean LDL-C are those carrying more around the middle than their training suggests, those with triglycerides above about 1.7 mmol/L, those with prediabetes or type 2 diabetes, and those already on a statin with a tidy LDL-C. If that is you, an HbA1c reading belongs in the same conversation.

Getting the number in the UK

NICE guideline NG238, published in December 2023, runs UK primary care on non-HDL cholesterol. Not LDL-C, and not apoB.

The committee had a good reason. Non-HDL-C needs no fasting sample and no triglyceride ceiling, and it is calculated from a test every laboratory already runs. It is also a decent proxy — 34% against 43% in the 2011 meta-analysis.

Decent is not the same as equivalent. Where apoB does appear in UK practice is the enhanced lipid profile, set out in a 2025 consensus from HEART UK and the Association for Laboratory Medicine. ApoB and lipoprotein(a) both sit on it.

Practically, ask your GP whether an enhanced profile is indicated for you, particularly with raised triglycerides or a family history. Failing that, apoB appears on most private cardiovascular panels, usually bundled with Lp(a), for roughly £30 to £60.

No fasting needed. And the assay is directly standardised, which calculated LDL-C is not.

What moves apoB, and what does not

Here our own category comes off badly, so let us be direct about it.

Fish oil is genuinely good at lowering triglycerides. It is much less good at lowering particle count. REDUCE-IT gave 4 g of icosapent ethyl daily to 8,179 statin-treated patients and reported a 19.7% placebo-corrected fall in triglycerides. ApoB fell 9.7%.

That was a prescription dose, roughly five times what a normal capsule delivers. At ordinary supplement doses in people with high triglycerides, meta-analyses have found no significant change in apoB at all, alongside a small rise in LDL-C.

A 2023 dose-response meta-analysis of 90 randomised trials and 72,598 participants found triglycerides and non-HDL-C falling close to linearly with omega-3 dose. LDL-C and HDL-C traced J-shaped curves instead.

Ference's data explains the arithmetic. Shifting apoB by a given amount through triglyceride lowering takes roughly five times the change you would need through LDL-C lowering. Triglyceride-rich particles carry a lot of triglyceride per particle, so moving the mass moves the count only slowly.

So omega-3 earns its place for dietary intake and triglycerides. Buying it to shift your particle count is not supported by the trials, and we are not going to suggest otherwise. If you want the fuller picture on evidence quality, we wrote about reading study results separately.

What reliably moves apoB is what moves LDL receptor activity and remnant clearance: lipid-lowering medication, bodyweight, alcohol intake and the saturated-fat share of the diet. The first of those is a GP conversation, not a supplement one.

What to do with the number

If your LDL-C is unremarkable and you have no metabolic red flags, apoB will probably tell you what you already know. Spend the £40 elsewhere.

Order it if your triglycerides run high, if your waist has crept up, if you have prediabetes or diabetes, if there is early heart disease in the family, or if you are on a statin and want to know whether the residual risk is real. Those are the situations where the two numbers part company.

Read it against a threshold, not a population average. The 2019 ESC/EAS goals are under 1.0 g/L at moderate risk, under 0.8 g/L at high risk and under 0.65 g/L at very high risk. Which band applies to you is a clinical judgement, so take the result to someone who can make it.

Then track the direction. One apoB is a snapshot; three over four years is information. Pair it with hs-CRP and HbA1c and you have a fair picture of where your risk is heading.

And keep the hierarchy straight. LDL-C is a useful, free, universally available estimate of something else. ApoB is the thing itself.

References

1. Sniderman AD, Williams K, Contois JH, et al. (2011). "A meta-analysis of low-density lipoprotein cholesterol, non-high-density lipoprotein cholesterol, and apolipoprotein B as markers of cardiovascular risk." Circulation: Cardiovascular Quality and Outcomes, 4(3), 337-345. doi:10.1161/CIRCOUTCOMES.110.959247.
2. Richardson TG, Sanderson E, Palmer TM, et al. (2020). "Evaluating the relationship between circulating lipoprotein lipids and apolipoproteins with risk of coronary heart disease: a multivariable Mendelian randomisation analysis." PLOS Medicine, 17(3), e1003062. doi:10.1371/journal.pmed.1003062.
3. Ference BA, Kastelein JJP, Ray KK, et al. (2019). "Association of triglyceride-lowering LPL variants and LDL-C-lowering LDLR variants with risk of coronary heart disease." JAMA, 321(4), 364-373. doi:10.1001/jama.2018.20045. PMID 30694319.
4. Marston NA, Giugliano RP, Melloni GEM, et al. (2022). "Association of apolipoprotein B-containing lipoproteins and risk of myocardial infarction in individuals with and without atherosclerosis: distinguishing between particle concentration, type, and content." JAMA Cardiology, 7(3), 250-256. doi:10.1001/jamacardio.2021.5083. PMID 34773460.
5. Sniderman AD, Dufresne L, Pencina KM, et al. (2024). "Discordance among apoB, non-high-density lipoprotein cholesterol, and triglycerides: implications for cardiovascular prevention." European Heart Journal, 45(27), 2410-2418. doi:10.1093/eurheartj/ehae258.
6. Solnica B, Sniderman AD, Wyszomirski A, et al. (2023). "Concordance/discordance between serum apolipoprotein B, low density lipoprotein cholesterol and non-high density lipoprotein cholesterol in NATPOL 2011 participants." International Journal of Cardiology, 390, 131150. doi:10.1016/j.ijcard.2023.131150.
7. Kenkre JS, Mazaheri T, Neely RDG, et al. (2025). "Standardising lipid testing and reporting in the United Kingdom; a joint statement by HEART UK and The Association for Laboratory Medicine." Annals of Clinical Biochemistry. doi:10.1177/00045632251315303.
8. National Institute for Health and Care Excellence (2023). "Cardiovascular disease: risk assessment and reduction, including lipid modification." NICE guideline NG238, published 14 December 2023.
9. Wang T, Zhang X, Zhou N, et al. (2023). "Association between omega-3 fatty acid intake and dyslipidemia: a continuous dose-response meta-analysis of randomized controlled trials." Journal of the American Heart Association, 12(11), e029512. doi:10.1161/JAHA.123.029512.
10. Bhatt DL, Steg PG, Miller M, et al. (2019). "Cardiovascular risk reduction with icosapent ethyl for hypertriglyceridemia (REDUCE-IT)." New England Journal of Medicine, 380(1), 11-22. doi:10.1056/NEJMoa1812792.
11. Mach F, Baigent C, Catapano AL, et al. (2020). "2019 ESC/EAS Guidelines for the management of dyslipidaemias: lipid modification to reduce cardiovascular risk." European Heart Journal, 41(1), 111-188. doi:10.1093/eurheartj/ehz455.

Omega-3 with the numbers on the label

Our algal oil gives 500 mg DHA and 250 mg EPA per daily serving, with third-party TOTOX and heavy metal results on every batch. It is a triglyceride and dietary-intake product, not a particle-count one — this article explains the difference.

Shop Omega-3 DHA →