Lipoprotein(a) Test

Your standard cholesterol panel leaves a blind spot. It checks LDL, HDL, and triglycerides, but it completely ignores a particle that independently drives heart attacks, strokes, and even heart valve disease. That particle is Lp(a) (lipoprotein(a)), and roughly one in five people carries enough of it to significantly raise their cardiovascular risk without ever knowing it.

What makes Lp(a) different from every other lipid marker is that your level is almost entirely set by your DNA. Diet and exercise barely move it. If your number is high, it has probably been high your entire adult life, quietly adding risk on top of whatever your other cholesterol numbers show. The 2026 ACC/AHA guidelines now recommend that every adult get tested at least once.

What Lp(a) Is and Why It Is Dangerous

Lp(a) looks a lot like an LDL particle. It has the same cholesterol-filled core and the same ApoB protein on its surface. But Lp(a) also carries a second protein called apolipoprotein(a), which is physically linked to the ApoB by a chemical bond. This extra protein is what gives Lp(a) its unique, dangerous properties.

Apolipoprotein(a) closely resembles plasminogen, a protein your body uses to dissolve blood clots. Because of that resemblance, Lp(a) can interfere with clot breakdown, essentially making your blood more prone to clotting. On top of that, Lp(a) particles carry large amounts of oxidized fats on their surface that trigger inflammation in artery walls. So Lp(a) attacks your cardiovascular system from multiple angles: it deposits cholesterol, it promotes inflammation, and it impairs your natural clot-dissolving machinery.

Your liver is the sole source of Lp(a) production. How much you produce depends primarily on the LPA gene, which controls the size and production rate of the apolipoprotein(a) protein. Smaller versions of this protein are produced more efficiently, which is why people who inherit genes coding for smaller apolipoprotein(a) tend to have higher Lp(a) levels. More than 90% of the variation in Lp(a) between people is genetic.

Heart Attack and Coronary Disease

Heart attack is the outcome most strongly linked to elevated Lp(a). People with levels in the top 5% of the population face 1.7 to 3 times higher risk of heart attack compared to those with low levels. In a pooled analysis of over 27,000 U.S. adults from five major community-based studies followed for an average of 21 years, those at or above the 90th percentile for Lp(a) had a 46% higher risk of atherosclerotic cardiovascular events compared to those below the 50th percentile (HR 1.46, 95% CI: 1.33 to 1.59).

The UK Biobank, which tracked over 460,000 people for a median of 11.2 years, confirmed a continuous, linear relationship: for every 50 nmol/L increase in Lp(a), the risk of atherosclerotic cardiovascular disease rose by about 11%. People with very high levels (at or above 150 nmol/L) had roughly 50% higher risk than those with low levels. This relationship held across White, South Asian, Black, and Chinese participants.

These risks persist even when LDL cholesterol is well controlled. A participant-level analysis across multiple trials showed that among people who achieved very low LDL cholesterol on statin therapy (as low as 3 to 77 mg/dL), those with Lp(a) above 50 mg/dL still had 38% higher cardiovascular risk than those with lower Lp(a). The two risk factors are independent and additive: the greatest danger comes when both Lp(a) and LDL cholesterol are elevated (90% increased risk), but high Lp(a) alone remains dangerous even with optimal LDL control.

Aortic Valve Disease

Lp(a) is unique among lipid markers in its connection to calcific aortic valve stenosis, a condition where the heart's aortic valve stiffens and narrows due to calcium deposits. A meta-analysis of 44 studies found that people with high Lp(a) had aortic valve disease at levels averaging 22.63 nmol/L higher than people without the condition. Those with elevated Lp(a) also experienced faster disease progression and faced about 39% higher risk of adverse outcomes including death (HR 1.39, 95% CI: 1.01 to 1.90).

No other cholesterol particle is as clearly linked to valve disease. If you have a family history of aortic stenosis or if you have been told you have early valve calcification, knowing your Lp(a) level adds context that standard lipid testing cannot provide.

Stroke, Peripheral Artery Disease, and Heart Failure

The cardiovascular effects of Lp(a) extend beyond the coronary arteries. People with levels in the top 5% face 1.7 to 3 times higher risk of peripheral artery disease (reduced blood flow to the legs) and heart failure. The association with stroke is more modest but still significant, with risk elevations of 1.2 to 1.6 times for those with the highest levels.

In the same multi-ethnic U.S. pooled cohort (mean follow-up of 21 years), elevated Lp(a) at or above the 90th percentile was also associated with higher risks of heart failure and stroke compared to those below the 50th percentile. These associations remained significant after adjusting for age, sex, race, blood pressure, BMI, LDL cholesterol, HDL cholesterol, triglycerides, smoking, diabetes, and family history.

Reference Ranges

Lp(a) is reported in either milligrams per deciliter (mg/dL) or nanomoles per liter (nmol/L). About 70% of U.S. labs use mg/dL, while the rest use nmol/L. An approximate conversion is: nmol/L equals 2.18 times the mg/dL value minus 3.83, though this varies somewhat depending on the lab's specific assay. Different assays can produce different numbers from the same blood sample, particularly for people who carry unusually large or small versions of the apolipoprotein(a) protein. For the most accurate comparisons over time, get your tests run by the same lab using the same method.

These tiers are drawn from multiple major guidelines including the 2026 ACC/AHA dyslipidemia guidelines, the European Atherosclerosis Society, and the American Heart Association. Your lab may use slightly different cutpoints. The key threshold recognized across all guidelines is 50 mg/dL (125 nmol/L), above which Lp(a) is considered a risk-enhancing factor. Risk accelerates continuously above roughly 24 to 30 mg/dL.

Ancestry and Sex Differences

Lp(a) levels vary by ancestry. In the UK Biobank, median levels were about 75 nmol/L in people of African ancestry, 31 nmol/L in South Asian individuals, 19 nmol/L in White individuals, and 16 nmol/L in East Asian individuals. Despite these differences in absolute levels, the proportional increase in cardiovascular risk per unit of Lp(a) is remarkably similar across all groups. Most guidelines therefore recommend the same thresholds regardless of ethnicity.

Women tend to have levels about 17% higher than men after age 50, likely related to the hormonal changes of menopause. If you had your Lp(a) measured before menopause, the ACC/AHA guidelines suggest getting retested about five years after menopause or after age 50 to see whether your level has shifted.

Tracking Your Trend

Unlike most blood biomarkers, Lp(a) is remarkably stable in the same person over time. In large studies, repeat measurements taken years apart correlate almost perfectly (correlation coefficient of 0.96 to 0.97), which is far stronger than repeat measurements of LDL cholesterol (0.70) or triglycerides (0.63). A Korean study of over 230,000 adults confirmed excellent reproducibility with an intraclass correlation above 0.99. Because levels are genetically locked in, a single measurement is usually enough to know where you stand.

There is one exception worth knowing: if your initial result falls in the borderline range (roughly 30 to 50 mg/dL), your risk classification is less certain. Studies show that 35% to 58% of people with borderline levels shift into a higher or lower risk category on repeat testing. Factors associated with greater Lp(a) change over time include Black race, female sex, diabetes, high blood pressure, elevated total cholesterol, and statin therapy. If your first reading is borderline, getting a second measurement several months later can clarify whether you are truly in the elevated range.

If you start a therapy that targets Lp(a) (such as a PCSK9 inhibitor or, in the future, one of the RNA-based therapies currently in clinical trials), repeat testing becomes essential to confirm that the treatment is working and to gauge its magnitude. Test again 8 to 12 weeks after starting treatment, and then at least annually.

When Results Can Be Misleading

Although Lp(a) is one of the most stable biomarkers you can test, certain conditions can temporarily push your number in one direction or the other. The most important confounder is acute illness or major inflammation. The LPA gene contains an inflammation-response element, so events like heart attack, severe infection, sepsis, or major surgery can shift Lp(a) dramatically. After a heart attack, Lp(a) may initially drop, then spike to as much as three times baseline within one to two weeks, and take three to six months to settle back to its true level. During severe illness like sepsis, Lp(a) can fall to as little as one-fifteenth of its usual value.

If you have been seriously ill, had surgery, or experienced a major cardiovascular event within the past three to six months, your Lp(a) result may not reflect your true genetic baseline. Wait until you have fully recovered before testing.

• Kidney disease: can raise Lp(a) levels in a stage-dependent manner, particularly in people with larger apolipoprotein(a) variants. The increase can appear even before kidney function tests like GFR become overtly abnormal.
• Liver disease: tends to lower Lp(a) levels as liver function declines, since the liver is the sole production site.
• Thyroid disease: hypothyroidism (underactive thyroid) may raise Lp(a), while hyperthyroidism (overactive thyroid) may lower it.
• Assay variability: different lab methods can produce different numbers from the same sample, especially for people whose apolipoprotein(a) protein is unusually large or small. One recent clinical study found that the within-person coefficient of variation for Lp(a) was about 19%, higher than the 10.2% that older reference databases suggested. For the most reliable tracking, use the same lab and same assay over time.

Fasting is not required for Lp(a) testing, and there is no meaningful difference between fasting and non-fasting results. One study found median fasting levels of 17.3 mg/dL versus 19.4 mg/dL at three to four hours after eating, a difference that was not statistically significant. Lp(a) has no circadian rhythm, so the time of day does not matter either.