Lipoprotein(a), pronounced "L-P little a," is an inherited cardiovascular risk factor. Roughly one in five people have elevated Lp(a), and most never find out — it is not on standard panels and it does not change with diet, exercise, or standard lipid medication. A single test establishes your lifetime baseline.
The Verdict
What Lp(a) physically measures
A note on naming before anything else, because it causes real confusion when people try to look up their own result. This marker is written Lp(a), LPA, or lp a depending on where you encounter it, and it is spoken as "lipoprotein little a" or simply "L-P-little-a". They all refer to the same thing: lipoprotein(a), the particle measured by this test. The one distinction worth keeping straight is that LPA in italics is also the name of the gene that determines how much of the particle you produce — so a laboratory report showing "LPA" is reporting the blood level, while a genetic report mentioning LPA is describing the gene behind it. It is also frequently confused with apo(a), which is the distinctive protein attached to the particle rather than the particle itself, and with ApoB — related, but not the same thing. Every Lp(a) particle carries one ApoB molecule, so Lp(a) is already counted inside a total ApoB result; what ApoB cannot tell you is how much of that particle count is Lp(a), which is why the two are measured separately.
An Lp(a) particle is an LDL particle with an extra protein, apolipoprotein(a), stitched onto it. That extra protein does two harmful things at once. It carries oxidized phospholipids that provoke inflammation in the artery wall, and it structurally resembles plasminogen, which interferes with the body's ability to break down clots.
So Lp(a) is both atherogenic and prothrombotic. It drives plaque and it makes a clot on that plaque more likely to persist. It is also the main lipid marker linked to calcific aortic valve stenosis, which is a distinct disease from coronary artery disease.
Standard reporting vs risk thresholds
There is no single "normal range" printed the same way across labs, partly because two incompatible units are in circulation. Both are listed below.
| Category | Mass units | Molar units | What it means |
|---|---|---|---|
| Desirable | Under 30 mg/dL | Under 75 nmol/L | No meaningful added cardiovascular risk from this marker. |
| Grey zone | 30–50 mg/dL | 75–125 nmol/L | Modest added risk. Weigh it against the rest of the picture. |
| Elevated | Above 50 mg/dL | Above 125 nmol/L | An independent risk factor. Most guidelines treat this as the action threshold. |
| High | Above 70 mg/dL | Above 180 nmol/L | Substantially raised lifetime risk of heart attack, stroke, and aortic stenosis. |
| Very high | Above 100 mg/dL | Above 250 nmol/L | Roughly the top 5%. Warrants specialist lipid input and family screening. |
Why lifestyle does not lower it
Lp(a) concentration is set almost entirely by the LPA gene, specifically by how many kringle IV type 2 repeats you inherited. Fewer repeats means a smaller apo(a) isoform, faster production, and higher blood levels. That is fixed at conception.
Diet, weight loss, exercise, and statins do not meaningfully change the number. This is biology, not a failure of effort. The value of the test is knowing your number, not chasing it downward.
The few things that do move Lp(a)
| Intervention | Effect on Lp(a) | Notes |
|---|---|---|
| Statins | No reduction; may raise it ~10% | Still indicated — they lower ApoB, which is the lever you actually have. |
| Diet, exercise, weight loss | Essentially no effect | Lp(a) is roughly 90% genetically determined. |
| Niacin | Lowers 20–25% | Abandoned for this purpose — outcome trials showed no benefit and real harms. |
| PCSK9 inhibitors | Lower 20–30% | A useful secondary effect on top of substantial ApoB reduction. |
| Menopausal hormone therapy | Lowers modestly | A real effect, but not a reason on its own to start or continue therapy. |
| Lipoprotein apheresis | Lowers 60–70% acutely | Reserved for severe, progressive disease with very high Lp(a). |
| RNA-targeted therapies in trials | Lower 80%+ | Pelacarsen, olpasiran, and lepodisiran. Cardiovascular outcome trials are ongoing, not complete. |
How and when to test
- Fasting is not required. Lp(a) is unaffected by meals.
- Ask for molar units (nmol/L) if available, since mass assays are biased by apo(a) isoform size.
- Do not test during an acute illness. Lp(a) behaves as an acute-phase reactant and rises with inflammation, infection, and after a heart attack. Wait 6–8 weeks.
- Re-test after a change in kidney function. Nephrotic syndrome can raise Lp(a) markedly, and it falls again when the condition is treated.
- Check thyroid status if a result is unexpectedly high. Untreated hypothyroidism raises Lp(a), and treatment lowers it.
- Test earlier rather than later. The result is most useful when there are decades left in which to act on it.
What to do if Lp(a) is elevated
- Tighten the ApoB target. When one risk factor cannot be lowered, the standard response is to lower every other one further — usually to an ApoB under 65 mg/dL.
- Consider a coronary calcium score. It converts an inherited statistical risk into a measurement of what has actually happened in your arteries.
- Screen first-degree relatives. Lp(a) is autosomal codominant, so parents, siblings, and children each have roughly a 50% chance of sharing the elevation.
- Control blood pressure and stop smoking. Both multiply Lp(a)-associated risk rather than merely adding to it.
- Ask about echocardiography over time if there is any murmur, given the aortic valve association.
- Do not start niacin for this. It lowers the number without improving outcomes, and it carries real side effects.
What it means in combination
- High Lp(a) + familial hypercholesterolemia: the highest-risk lipid combination in clinical practice. This belongs with a lipid specialist.
- High Lp(a) + high ApoB: two atherogenic exposures stacked. Aggressive ApoB lowering is the available lever.
- High Lp(a) + high hsCRP: particles plus an inflamed vessel wall. Lp(a) carries oxidized phospholipids, so the two are mechanistically linked.
- High Lp(a) + a calcium score of zero at 55: genuinely reassuring in the near term. It supports continued prevention rather than escalation.
- High Lp(a) + LDL-C stuck above target on a maximal statin: the Lp(a) cholesterol contribution described above. Correct the target rather than the drug.
- High Lp(a) + a family history of early heart attack: often the explanation for that history. Test the relatives.
When a result warrants seeing a physician
- Lp(a) above 125 nmol/L (50 mg/dL), particularly under age 50 when there is the most time to act.
- Any result above 250 nmol/L, which warrants specialist lipid input and family cascade screening.
- Elevated Lp(a) with a first-degree relative who had a heart attack or stroke before 55 in men or 65 in women.
- Elevated Lp(a) with a known heart murmur or diagnosed aortic valve disease.
- Any chest pain, exertional breathlessness, or new exercise intolerance — that is urgent and independent of the number.
Frequently Asked Questions
What is a good Lp(a) level?
Under 30 mg/dL, or under 75 nmol/L, is desirable. Above 50 mg/dL (125 nmol/L) is the threshold most guidelines treat as an independent cardiovascular risk factor. Above 180 nmol/L is high, and above 250 nmol/L puts you roughly in the top 5% of the population.
Are mg/dL and nmol/L interchangeable?
No, and this trips people up constantly. The two units measure different things — mass versus particle number — and there is no reliable conversion factor between them. Mass assays are also distorted by apo(a) isoform size, which varies between people. If you have a choice, ask for a molar (nmol/L) result, and always compare a repeat test in the same units from the same lab.
Can I lower Lp(a)?
Not meaningfully with lifestyle. Lp(a) is roughly 90% inherited and does not respond to diet, exercise, or weight loss. Statins do not lower it and may raise it slightly. PCSK9 inhibitors lower it 20–30%. RNA-targeted drugs in trials achieve reductions above 80%, but their cardiovascular outcome trials have not yet reported.
How often should I test Lp(a)?
Once for most people. Lp(a) is largely genetic and stable across adult life, so a single measurement establishes your lifetime baseline. Re-test if kidney function changes markedly, if thyroid disease is diagnosed, if you were tested during an acute illness, or if a new Lp(a)-lowering therapy becomes available to you.
Why is Lp(a) not on standard panels?
Historically because nothing could be done about it, and a test with no treatment attached is hard to justify. That reasoning is weakening. Knowing an elevated Lp(a) changes how aggressively ApoB is treated, when imaging is ordered, and whether relatives get screened — all actionable today. Major consensus statements now recommend measuring it at least once in every adult.
Does elevated Lp(a) differ by ancestry?
Yes, substantially. Median Lp(a) levels are considerably higher in people of African ancestry and lower in East Asian populations, with South Asian populations intermediate. Much of the risk-threshold evidence comes from predominantly European cohorts, so how those exact cut-points transfer across populations is still being worked out. The direction of risk holds; the precise numbers are debated.