Most cardiovascular risk factors are things you accumulate. Blood pressure creeps up. LDL responds to diet and statins. Weight and smoking are choices with consequences.
Lipoprotein(a) is not like that. It is roughly 90% genetically determined, essentially fixed from childhood, and unmoved by diet, exercise, or standard cholesterol drugs. About one in five people worldwide carries enough of it to raise cardiovascular risk, and most of them have never been told.
For the first time, a US guideline now recommends that every adult have it measured once.
A Number Set Before You Were Born
Lp(a) is an LDL-like particle with an extra protein, apolipoprotein(a), covalently attached. The gene that codes for it, LPA, contains a stretch called the kringle IV type 2 region that varies in copy number between individuals.
Copy number is the main lever. Fewer repeats produce smaller apo(a) isoforms and higher circulating Lp(a). A review in Current Cardiovascular Risk Reports notes that smaller isoforms have been associated with roughly double the risk of coronary heart disease and ischemic stroke, and that Lp(a) may be several times more atherogenic than LDL cholesterol particle for particle.
The result is a biomarker unlike anything else on a lipid panel. Unlike LDL, which is roughly normally distributed, Lp(a) is highly skewed across the population. Because levels remain stable throughout life, a single measurement usually suffices.
The particle contributes to arterial disease through three routes: depositing cholesterol in the artery wall as LDL does, promoting inflammation as the main carrier of oxidized phospholipids, and interfering with clot breakdown, since apo(a) resembles plasminogen but cannot be activated.
Why a Normal LDL Does Not Rule It Out
This is the part that catches people out. A patient can hit every LDL target and still carry the risk.
A participant-level meta-analysis of 27,658 people enrolled in six placebo-controlled statin trials tested exactly that. Among statin-treated participants, those with Lp(a) above 50 mg/dL showed elevated cardiovascular risk across every quartile of achieved LDL cholesterol. Even in the group with the lowest achieved LDL, below roughly 77 mg/dL, the hazard ratio was 1.38. The greatest risk appeared when both Lp(a) and LDL sat in the highest range, at 1.90.
Lp(a) operates through pathways distinct from LDL cholesterol, and lowering one does not address the other. Someone with a clean panel and a family history of early heart attacks may be looking at a risk factor their panel was never designed to detect.
Ancestry Shifts the Baseline, Not the Risk Per Unit
Because the driver is a gene, average levels differ by ancestry, and the pattern is documented.
One large analysis cited in that same review found a median Lp(a) of 75 nmol/L in Black participants, compared with 31 nmol/L in South Asian participants, 19 nmol/L in White participants, and 16 nmol/L in East Asian participants. Postmenopausal women also tend to have higher rates than men.
One point matters more than the averages: while concentrations vary across ethnic groups worldwide, the associated cardiovascular risk remains similar across populations. Elevated Lp(a) is equally atherogenic wherever it appears. What differs is how many people in a population carry it, which is an argument for testing everyone rather than for ancestry-specific thresholds.
What Changed in 2026, and What Did Not
The 2026 ACC/AHA multisociety dyslipidemia guideline gives universal one-time Lp(a) testing in adults a Class I recommendation, the strongest designation available. The full document in Circulation also restores LDL treatment goals scaled to risk, adds selective apolipoprotein B measurement, expands coronary artery calcium scoring, and replaces the Pooled Cohort Equations with the PREVENT equations. A guideline-at-a-glance summary in JACC sets the changes against the 2018 version. Guidance now extends across the lifespan, with special considerations for testing people under 18.
The threshold flagged is 125 nmol/L, approximately 50 mg/dL, corresponding to the 80th population percentile. Below that, between roughly 75 and 125 nmol/L, is generally treated as intermediate.
What has not changed is treatment. There is still no approved drug that lowers Lp(a). Four candidates are in late-stage outcome trials, with the first cardiovascular results expected before the end of 2026, and none have yet shown that lowering Lp(a) prevents heart attacks or strokes. Current management means treating everything else more aggressively.
Because the trait is inherited, a high result carries implications for siblings, parents, and children, and cascade testing of first-degree relatives is widely encouraged. That is a conversation for a clinician, not a reason to act on a number alone.
Key Questions Answered
How genetic is Lp(a)?
Roughly 90% of the variation is genetically determined, driven largely by copy number variation in the LPA gene. Levels remain stable throughout life and respond minimally to diet or exercise.
Why isn't it on my cholesterol results?
A standard lipid panel measures LDL, HDL, and triglycerides. Lp(a) requires a separate test order, and no US guideline recommended it universally until 2026.
Can I have high Lp(a) with normal LDL?
Yes. In a meta-analysis of six statin trials, people with Lp(a) above 50 mg/dL carried an elevated risk even in the lowest achieved LDL quartile.
Does ancestry affect my level?
Median levels differ by ancestry, with the highest reported in people of African descent. The risk associated with a given level is similar across groups.
How many times do I need the test?
Once, in most cases. Because levels are genetically fixed and stable, repeated measurement is generally unnecessary.
Is there a treatment?
Not yet. Four drugs are in outcome trials with first results expected in 2026, and none is approved. Elevated Lp(a) currently prompts more aggressive management of other risk factors.
