Lipoprotein(a): The Inherited Cardiovascular Marker
Roughly one in five people has an elevated level, it does not respond to diet or exercise, and most have never had it measured. A single test covers a lifetime.
The Short Answer
Lipoprotein(a) is unusual among cardiovascular markers in three ways that together make it worth a specific article. It is roughly 80 to 90 per cent genetically determined and stable through life, so one measurement suffices. It is an independent causal risk factor supported by genetic evidence rather than only observational association. And it is elevated in about one in five people, most of whom have never been tested for it.
What It Is
Lp(a) is an LDL-like particle with an additional protein, apolipoprotein(a), attached by a disulphide bond to apolipoprotein B. The apo(a) component has structural similarity to plasminogen, which is the basis for its proposed prothrombotic effects.
The particle is atherogenic in the same way LDL is, since it carries apoB and enters the arterial wall, and it carries oxidised phospholipids that add an inflammatory dimension. The plasminogen similarity suggests interference with fibrinolysis.
Concentration is determined largely by the LPA gene, particularly by the number of repeats in a structural domain: fewer repeats generally means smaller particles produced at higher concentration. That genetic determination is why the level is stable and why lifestyle does not move it.
Levels vary substantially by ancestry, with higher average concentrations in people of African descent and lower in some East Asian populations, and the relationship between concentration and risk appears to hold across groups.
The Evidence That It Is Causal
The case for causality rests on Mendelian randomisation, which uses inherited variants as a natural randomisation. Variants associated with lifetime higher Lp(a) are associated with higher cardiovascular incidence in a dose-dependent way, which is difficult to explain by confounding or reverse causation.
The associated conditions are atherosclerotic cardiovascular disease, including myocardial infarction and stroke, and calcific aortic valve stenosis, where the association is among the strongest for any modifiable-seeming factor.
Guideline positions have moved accordingly. Several major bodies now recommend measuring Lp(a) at least once in adults, with variation in how strongly, and it remains substantially under-measured in practice.
The size of the effect matters for framing. Elevated Lp(a) raises risk meaningfully and does not determine outcome, and a person with high Lp(a) and otherwise excellent risk factors is in a different position from one with high Lp(a) and untreated hypertension and elevated apoB.
Measuring It Once, Properly
| Consideration | Detail |
|---|---|
| Units | nmol/L is preferred over mg/dL; the two are not reliably interconvertible |
| Frequency | Once in a lifetime is generally sufficient, given genetic determination |
| Fasting | Not required |
| Assay variation | Some assays are affected by isoform size; modern assays address this |
| Thresholds | Commonly cited concern above roughly 125 nmol/L, with risk continuous rather than threshold-based |
| Situations that alter it | Acute inflammation, pregnancy, kidney disease and thyroid dysfunction can shift it temporarily |
| Cascade testing | First-degree relatives share the genetics; testing them is reasonable |
The units issue causes real confusion, since a result in mg/dL and one in nmol/L cannot be compared reliably, and conversion factors are approximate. Recording the units alongside the value is necessary for any future comparison.
The cascade testing point is the one most often missed. Because it is inherited, an elevated result in one person means first-degree relatives are more likely to be elevated, and telling them is a concrete action available immediately.
What Does Not Lower It
Being clear about this saves effort and money.
Diet. No dietary pattern meaningfully lowers Lp(a). Saturated fat reduction lowers LDL and can slightly raise Lp(a) in some studies, which is a curious and clinically unimportant finding.
Exercise. No meaningful effect.
Weight loss. No meaningful effect.
Statins. No reduction, and slight increases have been reported. This does not argue against statins, whose apoB reduction remains beneficial.
Supplements. None with credible evidence. High-dose niacin lowers Lp(a) modestly and did not improve outcomes in trials and has adverse effects, so it is not used for this purpose.
What does lower it: PCSK9 inhibitors reduce it by roughly a quarter as a secondary effect, and lipoprotein apheresis lowers it substantially and is used in specific severe cases. Neither is prescribed for Lp(a) alone.
What is coming: antisense oligonucleotides and small interfering RNA agents targeting apo(a) production lower it dramatically, by 80 per cent or more, and are in late-stage outcome trials. Whether that reduction translates into fewer events is the question those trials are answering, and it is among the most consequential pending results in cardiovascular medicine.
What to Do With an Elevated Result
Since the level itself is not modifiable, the response is to reduce everything else that compounds it.
Apolipoprotein B, aggressively. Lp(a) is one atherogenic particle among many, and reducing the rest lowers total particle burden. Guideline targets for apoB and LDL are lower for people at higher risk, and elevated Lp(a) is a reason to sit at the more aggressive end. This usually means a clinical conversation about medication.
Blood pressure, to target.
Smoking, absolutely not.
Glycaemic control.
Consider a coronary calcium score, where it would change a decision, since it measures disease rather than risk.
Tell first-degree relatives. Concrete, free and frequently omitted.
Aortic valve awareness. The association with calcific aortic stenosis means a murmur or exertional symptoms warrant assessment rather than reassurance.
Do not: spend money on supplements for it, or read the result as fatalistic. The modifiable factors do more in someone with high Lp(a) than in someone without, because the baseline risk they are acting on is higher.
Why It Is Under-Measured
The historical reason is straightforward: for decades there was nothing to do about it, so measuring it produced anxiety without an action. Clinical practice reasonably deprioritised a test whose result did not change management.
That reasoning has weakened. Knowing the level changes how aggressively the modifiable factors are pursued, it identifies families for cascade testing, and it informs interpretation of otherwise unremarkable lipid panels in someone with premature disease. Guidelines have moved, and practice has lagged.
If targeted therapies succeed in outcome trials, the calculus changes entirely and everyone with elevated Lp(a) becomes a candidate for a specific intervention. Knowing your level in advance of that is a reasonable position to be in, and it costs one test.
The practical instruction is simple: ask for it once, record the value and the units, and act on the rest.
The AEONNN Perspective
Lp(a) is the clearest example in AEONNN's model of a fixed factor that changes the weighting of everything modifiable. It is roughly 80 to 90 per cent genetically determined, stable through life and unresponsive to diet, exercise, weight loss or statins, so the platform reads it once and uses it as calibration rather than as a target.
What it calibrates is urgency. Elevated Lp(a) means Pillar 4's modifiable factors, apolipoprotein B, blood pressure, smoking and glycaemic control, carry more weight, because the baseline they act on is higher. That is the opposite of fatalism and it is the practical use of a fixed marker.
The Quality layer carries the units problem, since nmol/L and mg/dL are not reliably interconvertible and a value without its units cannot be compared later. The Population layer carries the ancestry variation. And the platform surfaces two actions people miss: telling first-degree relatives, who share the genetics, and awareness of the calcific aortic stenosis association. The Innovation layer tracks the apo(a)-targeting agents in late-stage outcome trials, which are among the most consequential pending results in cardiovascular medicine.
Pillar Matrix mapping
Database Matrix layers
- Evidence Layer (PubMed, Cochrane, ClinicalTrials.gov)
- Meta / Consensus Layer (JAMA, BMJ, specialty society positions)
- Population Layer (UK Biobank, NHANES)
- Innovation Layer (bioRxiv preprints, patent filings)
Frequently Asked
What is lipoprotein(a)?
An LDL-like particle with an additional apolipoprotein(a) attached. It is atherogenic like LDL, carries oxidised phospholipids and resembles plasminogen, which underlies proposed prothrombotic effects.
How often should Lp(a) be measured?
Once in a lifetime is generally sufficient, since it is roughly 80 to 90 per cent genetically determined and stable through life.
What level is concerning?
Concern is commonly cited above roughly 125 nmol/L, with risk continuous rather than threshold-based. Units matter: nmol/L is preferred and is not reliably interconvertible with mg/dL.
Can diet or exercise lower Lp(a)?
No. Neither diet, exercise nor weight loss meaningfully lowers it, and statins do not reduce it either, though their apoB reduction remains beneficial.
What should I do if mine is high?
Reduce everything else that compounds it: apolipoprotein B aggressively, blood pressure to target, no smoking, glycaemic control. Also tell first-degree relatives, who share the genetics.
Are there therapies coming?
Antisense oligonucleotide and small interfering RNA agents targeting apo(a) production lower it by 80 per cent or more and are in late-stage outcome trials.
Why has my doctor never measured it?
Historically there was no specific action available, so a test that did not change management was deprioritised. Guidelines have since moved, and practice has lagged.
Evidence and review
Any dosage ranges cited here reflect the ranges used in published human trials, not personal recommendations. Evidence in this field moves, so this article is reviewed quarterly and carries its last-updated date above. Nothing here is intended as medical advice, and supplementation should be discussed with a qualified clinician, particularly alongside prescribed medication or an existing condition.