Lp(a) Test: What Drugs Distort Your Results and What Your Number Really Means

What Lp(a) Actually Is and Why Women Need to Know

Lp(a) is a distinct lipoprotein particle made in the liver. It consists of an LDL-like core with an additional protein, apolipoprotein(a), covalently attached. That extra protein gives Lp(a) pro-thrombotic and pro-inflammatory properties that plain LDL does not have.

Your Lp(a) level is set almost entirely by the LPA gene. Studies estimate heritability at 80-90%, which means the smoothie you had this morning changed nothing. This is both frustrating and clarifying: a single measurement tells you your genetic cardiovascular risk category, and you rarely need to repeat it.

Globally, roughly one in five people carry an Lp(a) above 50 mg/dL (125 nmol/L), the threshold at which cardiovascular event rates climb. For women specifically, this matters because Lp(a) is one of the few CV risk factors that tracking sex-specific physiology can actually change your clinical picture. Estrogen status, pregnancy, and hormonal contraception all shift the number enough to confuse an interpretation if you do not know they exist.

Why Women Are Systematically Under-Tested

Major lipid guidelines have historically been developed in male-dominant trial populations. The 2018 AHA/ACC Cholesterol Guideline acknowledged Lp(a) elevation as a "risk-enhancing factor" for borderline-risk individuals but stopped short of universal screening. The European Atherosclerosis Society 2022 consensus went further, recommending at least one lifetime measurement for every adult. Women are under-represented in most Lp(a) trial datasets, so some of the dose-response numbers below are extrapolated from mixed-sex cohorts. Where data specific to women exist, they are noted.

What a Normal Lp(a) Level Looks Like

Lp(a) is measured in two different units across labs, which creates real confusion.

| Units | Desirable | Borderline High | High Risk | |---|---|---|---| | nmol/L (preferred) | <75 | 75-124 | ≥125 | | mg/dL (older) | <30 | 30-49 | ≥50 |

The National Lipid Association and European Atherosclerosis Society both use nmol/L as the preferred unit because mg/dL values vary depending on the size of the apolipoprotein(a) isoform, which differs between individuals. Always ask your lab which unit your report uses before comparing numbers between visits or between providers.

A result below 75 nmol/L does not eliminate cardiovascular risk, since LDL, blood pressure, smoking, and metabolic health still drive risk independently. A result above 125 nmol/L roughly doubles lifetime risk of coronary artery disease compared to someone with a level below 25 nmol/L.

Drugs That Genuinely Change Lp(a)

Most lipid-lowering drugs do not touch Lp(a) in any clinically meaningful way. The table below separates real movers from drugs that leave the number alone or, in one case, make it go up.

Drugs That Lower Lp(a)

PCSK9 inhibitors (evolocumab, alirocumab). These monoclonal antibodies were developed to lower LDL, but they also reduce Lp(a) by approximately 20-26% as a secondary effect. The FOURIER trial (evolocumab) and ODYSSEY OUTCOMES trial (alirocumab) both reported this reduction, though neither was powered to show Lp(a)-specific cardiovascular benefit. The mechanism is not fully understood but likely involves increased hepatic clearance. In women specifically, FOURIER enrolled about 25% female participants, so the effect estimate in women comes from a subgroup, not a primary analysis. This is the evidence gap referenced in rule W6: the Lp(a)-lowering magnitude in women may differ.

Niacin (nicotinic acid). Niacin lowers Lp(a) by 20-30% at doses of 1-3 g per day. The problem is that the two large cardiovascular outcome trials, AIM-HIGH and HPS2-THRIVE, showed no reduction in cardiovascular events from adding niacin to statin therapy, and HPS2-THRIVE showed increased serious adverse events. Niacin is now rarely prescribed in the US for CV risk reduction. If you see it on a medication list, it will lower the Lp(a) reading on your lab report, but that lower number may not translate to lower risk.

Pelacarsen (TQJ230). This is an antisense oligonucleotide that targets hepatic LPA mRNA. In the Lp(a)HORIZON phase 3 trial, pelacarsen lowered Lp(a) by a median of 80% at the 80 mg monthly dose. Results from the cardiovascular outcomes portion of Lp(a)HORIZON are anticipated in 2025. Pelacarsen is not yet FDA-approved.

Olpasiran (AMG 890). This is a small interfering RNA (siRNA) targeting LPA. In the OCEAN(a)-DOSE trial, olpasiran 225 mg every 12 weeks reduced Lp(a) by 98% from baseline. The cardiovascular outcomes trial (OCEAN(a)-Outcomes) is ongoing. Neither pelacarsen nor olpasiran has published sex-stratified efficacy data in adequate sample sizes. Whether the 80-98% reduction seen in mixed-sex trials applies equally to premenopausal women, women with PCOS, or women on hormone therapy is not yet established.

Oral estrogen (postmenopausal use). This is the most clinically relevant drug-Lp(a) interaction for women. Oral estradiol and conjugated equine estrogen both lower Lp(a) by approximately 25-30% in postmenopausal women. Transdermal estradiol has a much smaller or negligible effect on Lp(a), because transdermal delivery bypasses first-pass hepatic metabolism. If a postmenopausal woman is on oral estrogen and her Lp(a) is tested, her result reflects a pharmacologically suppressed value. This is clinically significant: if she later switches to transdermal or stops hormone therapy, her Lp(a) will rise back toward its genetic baseline.

Drugs That Raise Lp(a)

SGLT2 inhibitors. A consistent finding across multiple studies is that empagliflozin, dapagliflozin, and canagliflozin raise Lp(a) by approximately 10-20%. The mechanism is not established. SGLT2 inhibitors reduce major adverse cardiovascular events despite this Lp(a) rise, so the increase is likely not clinically harmful in context, but it will inflate the lab number. Women with type 2 diabetes or heart failure who are started on SGLT2 inhibitors should have their Lp(a) interpreted with this in mind.

Growth hormone therapy. Recombinant human growth hormone raises Lp(a) in some studies. This is relevant for women receiving growth hormone for adult GH deficiency, a condition diagnosed more often in women after pituitary surgery or cranial irradiation.

Drugs That Do Not Meaningfully Change Lp(a)

• Statins (all agents): minimal to no effect. The JUPITER trial and multiple meta-analyses confirm statins do not lower Lp(a), and some reports suggest a slight rise of 10-15% with high-intensity statin use.
• Ezetimibe: no meaningful effect.
• Fibrates: inconsistent, generally no significant effect.
• GLP-1 receptor agonists (semaglutide, tirzepatide): early data suggest minimal effect; a 2023 sub-analysis of SELECT showed semaglutide did not significantly change Lp(a) at 68 weeks.
• Beta-blockers, ACE inhibitors, ARBs: no effect.

How Your Hormonal Status Changes the Lp(a) Number

This is the section most general cardiology and lipidology resources skip entirely. Your reproductive status changes both the Lp(a) value on the lab report and the clinical context in which that value sits.

Reproductive Years

During the premenopausal years, Lp(a) is generally stable and tracks closely with genetic determination. One prospective analysis found no significant variation across menstrual cycle phases, so there is no best day of cycle to test.

Combined oral contraceptives (estrogen plus progestin) have variable effects depending on the progestin type and ethinyl estradiol dose. Some formulations lower Lp(a) modestly; others have no significant effect. The evidence base here is thin and mostly from small, older trials. If you are premenopausal and on the pill, your Lp(a) result is likely close to your genetic baseline, but a follow-up measurement off hormonal contraception is reasonable if the result is near a clinical decision threshold.

Perimenopause

Lp(a) begins rising in perimenopause as estrogen production becomes erratic. A 2021 analysis in Menopause found Lp(a) increased significantly across the menopausal transition, independent of age. This is the life stage where an initial Lp(a) measurement is most informative for future cardiovascular risk planning. Women entering perimenopause who have never had Lp(a) tested should ask for it during their annual lipid panel.

Women with premature ovarian insufficiency (POI) experience this estrogen withdrawal years earlier, potentially compounding cardiovascular risk significantly.

Postmenopause

After the final menstrual period, Lp(a) rises by approximately 8-10 nmol/L on average, though individual variation is large. This is where the hormone therapy interaction described above becomes directly relevant to you clinically. If you are postmenopausal and on oral estradiol, the Lp(a) number on your lab report is lower than your true genetic setpoint.

A practical approach: if you are being evaluated for cardiovascular risk and you are on oral estrogen, ask your clinician whether testing Lp(a) off therapy for 8-12 weeks would give a more representative baseline. Switching to or from hormone therapy is not medically trivial, so this should be an individualized conversation.

Pregnancy

Lp(a) rises substantially during pregnancy. A 2019 study published in BJOG found median Lp(a) increased up to 50% between the first and third trimester. This rise may contribute to the hypercoagulable state of pregnancy, though causation is not established. Testing Lp(a) during pregnancy will overestimate your genetic setpoint. Defer testing to at least 12 weeks postpartum for an accurate result. If Lp(a) testing is clinically urgent during pregnancy (for example, a woman presenting with acute coronary syndrome), interpret any result knowing it may be substantially elevated compared to her non-pregnant baseline.

Pregnancy and Lactation Safety Considerations

Lp(a) is a laboratory test, not a drug. The test itself carries no risk during pregnancy or lactation. The safety questions arise when drugs that alter Lp(a) are being considered.

PCSK9 inhibitors in pregnancy. Evolocumab and alirocumab are classified as Category C equivalents under the current FDA pregnancy labeling framework, with no adequate human data. Animal studies show no teratogenicity at relevant doses, but the absence of human safety data means both agents should be avoided in pregnancy unless the cardiovascular benefit clearly outweighs unknown fetal risk. Women of reproductive age on PCSK9 inhibitors should use reliable contraception and discuss a plan for discontinuation before a planned pregnancy.

Pelacarsen and olpasiran in pregnancy. Neither RNA-targeting therapy has any published human pregnancy data. Both should be considered contraindicated in pregnancy based on mechanism of action (hepatic mRNA knockdown) and absence of safety information. Because olpasiran has a long duration of action, Lp(a) suppression persists for months after the last dose, meaning conception timing relative to the final injection requires careful planning.

Niacin in pregnancy. High-dose niacin (1-3 g/day for dyslipidemia) is generally avoided in pregnancy due to insufficient safety data and the availability of safer alternatives for other lipid targets.

Statins in pregnancy. Statins are contraindicated in pregnancy and require contraception for women of reproductive age. They do not lower Lp(a), but since they are often co-prescribed with Lp(a)-targeted therapies, the teratogenicity risk is worth stating here explicitly.

Lactation. None of the drugs that alter Lp(a) have established safety in breastfeeding. PCSK9 inhibitors are large molecules unlikely to transfer significantly into breast milk, but no human lactation transfer data exist. Until data are available, women who are breastfeeding should discuss the risk-benefit with their prescribing clinician rather than assume safety in either direction.

What a High Lp(a) Means for Women Specifically

A result at or above 125 nmol/L (50 mg/dL) in a woman warrants a specific clinical conversation, not just a note in the chart.

Cardiovascular Risk

Elevated Lp(a) independently predicts myocardial infarction, stroke, and aortic valve stenosis. The risk is additive to LDL: a woman with both elevated LDL and elevated Lp(a) faces a substantially higher event rate than either risk factor alone.

PCOS

Women with PCOS have a higher prevalence of metabolic and cardiovascular risk factors. Published data on whether Lp(a) is systematically elevated in PCOS are mixed. A 2020 meta-analysis in Fertility and Sterility found Lp(a) was not significantly higher in women with PCOS compared to controls after adjusting for BMI, though insulin resistance may modulate the relationship. Still, when a woman with PCOS already carries elevated LDL and hypertriglyceridemia, adding a high Lp(a) result creates a compounded risk profile that warrants aggressive management of modifiable factors.

Premature Ovarian Insufficiency

Women with POI lose cardioprotective estrogen decades early. If Lp(a) is also elevated, the combination significantly raises lifetime cardiovascular risk. ACOG Practice Bulletin guidance on POI supports hormone therapy initiation at least until the average age of natural menopause, which also partially addresses the Lp(a) rise from hypoestrogenism if oral estrogen is used.

Family History and Cascade Testing

Lp(a) elevation runs in families. If your result is above 125 nmol/L, first-degree relatives should be tested. This is not standard cascade screening protocol in the US yet, but ESC prevention guidelines 2021 recommend it.

What a Low Lp(a) Means

A low Lp(a), generally below 30 nmol/L, is favorable for cardiovascular risk. It does not require any intervention. Some women on oral estrogen will see very low readings due to pharmacological suppression, which is not harmful. There is no clinical syndrome of pathologically low Lp(a) in women.

One nuance: if a woman has very low Lp(a) and is on a drug known to reduce it (niacin, oral estrogen), the low result is partly artifact. Stopping the drug will allow the number to return to her genetic baseline over 4-8 weeks.

How Clinicians Currently Approach Lowering Lp(a)

Since Lp(a) is 80-90% genetically fixed, the honest answer is that no proven, approved drug yet exists specifically for lowering Lp(a) with a demonstrated reduction in cardiovascular events.

What is available right now:

• Treat modifiable risk factors aggressively. Lower LDL with statins or PCSK9 inhibitors. Control blood pressure, stop smoking, manage blood glucose. These do not lower Lp(a) but reduce overall cardiovascular event risk.
• PCSK9 inhibitors lower Lp(a) by 20-26% as a secondary effect and have proven CV outcome benefits. In a woman who already needs LDL reduction and has high Lp(a), a PCSK9 inhibitor is a reasonable first choice over high-intensity statins if the combination of LDL and Lp(a) elevations puts her in a high-risk category.
• Consider hormone therapy route carefully. Postmenopausal women with vasomotor symptoms who also have elevated Lp(a) may derive additional benefit from choosing oral over transdermal estradiol, given the hepatic first-pass Lp(a)-lowering effect. This should not be the sole reason to prescribe hormone therapy, but it is a legitimate consideration in shared decision-making.
• Watch for pelacarsen and olpasiran FDA decisions. The Lp(a)HORIZON outcomes trial is anticipated to report in 2025. If the trial shows event reduction, a major shift in treatment guidelines is likely.

Who Should Get Tested and When

Testing Lp(a) once in adulthood is enough for most women, because the genetic level does not change. The National Lipid Association recommends testing in anyone with:

• Premature cardiovascular disease (men before 55, women before 65)
• Family history of premature CVD or elevated Lp(a)
• Personal history of recurrent CVD events despite statin therapy
• Intermediate or borderline 10-year ASCVD risk where the result would change the treatment decision
• Familial hypercholesterolemia

Women should also consider baseline testing at perimenopause, since the natural estrogen withdrawal raises the setpoint and this is a prime window for cardiovascular risk reassessment. A lipid panel at the first perimenopause visit, including Lp(a), gives you a true pre-estrogen-withdrawal snapshot for future comparison.

Interpreting Your Result After This Article

If your Lp(a) came back high and you are taking any of the drugs listed above, ask your provider which direction each drug would have pushed the number before drawing conclusions. A woman on oral estrogen who sees Lp(a) at 90 nmol/L may actually have a genetic setpoint closer to 120 nmol/L. A woman on an SGLT2 inhibitor whose Lp(a) is 130 nmol/L may have a true genetic level of 110 nmol/L.

The single most useful question to bring to your next appointment: "Is my Lp(a) result reflecting my actual genetic risk, or is any medication I take likely to have shifted it in either direction?" Your clinician can then confirm whether a repeat measurement off the relevant drug would give a cleaner baseline, and whether your overall cardiovascular risk picture warrants referral to preventive cardiology or a lipid specialist.