2019 NLA scientific statement: Lp(a) – key points

The various large meta-analyses, Mendelian randomizations, and prospective population-based studies have found that Lipoprotein (a) is an independent risk factor for atherosclerosis, aortic valve stenosis, and thrombosis. The Lp(a) test is considered elevated when its value is >50 mg/dL or >100 nmol/L, corresponding to the top 20th percentile of the general population.

Currently, there are no approved specific therapies for Lp(a). The National Lipid Association (NLA) does not recommend the use of niacin, hormone replacement therapy (HRT), or lomitapide (microsomal triglyceride transfer protein inhibitor). Recent trials such as FOURIER and ODYSSEY have shown that the addition of PCSK9 inhibitors to statin therapy can lower Lp(a) by 30%.

However, various guidelines, including the 2018 American Heart Association/American College of Cardiology (AHA/ACC) and the 2019 NLA scientific statement, recommend the use of PCSK9 inhibitors only in the context of uncontrolled LDLc/non-HDLc in patients at high risk or very high risk for ASCVD events.

Dr. Tashko

Also see:






Scientific Statement

April 2019


  • Apolipoprotein(a), attached to the Apo B segment of an LDL-like particle, is a unique protein contained within Lp(a).
  • Apo(a) has homology with plasminogen and may inhibit fibrinolysis, thus increasing thrombosis.
  • Through inhibition of fibrinolysis at sites of plaque rupture, Apo(a) has the potential to cause MI and ischemic stroke.
  • Thrombosis at sites of turbulent flow may promote atherosclerotic and valvular aortic stenosis.
  • Apo(a) possesses unique properties that promote initiation and progression of atherosclerosis and calcific valvular aortic stenosis through endothelial dysfunction and pro-inflammatory responses, and pro-osteogenic effects promoting calcification.
  • Many of these effects are likely attributable to the oxidized phospholipids, of which Lp(a) is the preferential carrier, and which are covalently attached to the apo(a) portion of Lp(a).


  • Meta-analyses of prospective, population-based studies of high Lp(a) demonstrate high risk of MI, CHD, and ischemic stroke.
  • Large prospective, population-based studies of high Lp(a) demonstrate high risk of MI, ischemic stroke, VAS, coronary artery stenosis, carotid stenosis, femoral artery stenosis, heart failure, cardiovascular mortality, and all-cause mortality.
  • Large Mendelian randomization and GWA studies confirm that high Lp(a) is a causal factor for MI, ischemic stroke, VAS, coronary artery stenosis, carotid stenosis, femoral artery stenosis, heart failure, cardiovascular mortality, and all-cause mortality.
  • These causal relationships are independent of concentrations of other lipids and lipoproteins, including LDLc.


  • Measurement of Lp(a) is currently not standardized or harmonized.
  • Available assays report Lp(a) in either mg/dL or nmol/L and may exhibit Lp(a) isoform-dependent bias.
  • Evidence is incomplete regarding the utility of using different cut points of Lp(a) based on age, gender, ethnicity, or the presence of comorbid conditions.
  • Elevated Lp(a) appears to confer elevated risk for ASCVD over a wide range of (“at any”) LDLc concentrations.
  • An Lp(a) level >50 mg/dL (>100 nmol/L) may be considered as a risk-enhancing factor favoring the initiation of statin therapy. This level corresponds to the top 20th percentile in populations which are predominantly Caucasian.
  • The corresponding top 20th population percentile in African Americans is approximately 150 nmol/L, but it is unclear whether a different risk threshold or cut-point should be applied.

    • Clinicians should be aware that African Americans have an approximately 3-fold higher median Lp(a) than Caucasian populations (75 nmol/L vs 20 nmol/L)

When to test

Lp(a) testing is reasonable to refine risk assessment for ASCVD events in adults with:

  • Family HX of 1st-degree relatives with premature ASCVD (<55 y of age in men; <65 y of age in women).
  • Personal HX of premature ASCVD (<55 y of age in men; <65 y of age in women).
  • Personal HX of severe hypercholesterolemia (LDLc ≥190 mg/dL) or suspected FH.

Lp(a) testing may be reasonable in adults:

  • To aid in the clinician-patient discussion about whether to prescribe a statin in those aged 40-75 with borderline (5%–7.4%) 10-y ASCVD risk.
  • To identify a possible cause for a less-than-anticipated LDLc lowering to evidence-based LDLc–lowering therapy.
  • To use in cascade screening of family members with severe hypercholesterolemia.
  • To identify those at risk for progressive VAS.


  • Lifestyle therapy, including diet and physical exercise, has no significant effect on Lp(a) levels.
  • Statin therapy does not decrease Lp(a) levels.
  • Patients with a history of ASCVD who are taking statins + have Lp(a) >50 mg/dL are at increased risk for ASCVD events, independent of other risk factors.
  • Niacin lowers Lp(a), has no demonstrated ASCVD risk reduction benefit in patients taking statins, and may cause harm.
  • Lomitapide, which is indicated to lower LDLc in patients with homozygous FH, also lowers Lp(a) but is not recommended for ASCVD risk reduction.
  • PCSK9 inhibitors addition lower Lp(a) by 30%, but the contribution of Lp(a) reduction to their ASCVD risk reduction benefit remains undetermined.
  • LDL aphaeresis lowers Lp(a) and is sometimes used for those with elevated Lp(a) + recurrent ASCVD events.
Lp(a) Statin

Super Key Points 1

  • The measurement of Lp(a) is reasonable in adults with:

    • Premature ASCVD (<55 of age in men, <65 of age in women).
    • Recurrent or progressive ASCVD, despite optimal lipid lowering.
  • Lp(a) is associated with an increased risk of calcific VAS proportional to the Lp(a) level, and measuring Lp(a) may be reasonable in patients with this disorder.
  • Patients with high Lp(a) levels may have less-than-expected LDLc lowering on statin therapy.
  • There is a LACK of current evidence demonstrating that lowering Lp(a), independently of LDLc, reduces ASCVD events in individuals with established ASCVD.

    • It appears that large absolute reductions in Lp(a) may be needed to demonstrate a significant clinical benefit.

Super Key Points 2

  • The LPA gene is fully expressed by 1-2 y of age and the concentration of Lp(a) reaches adult levels by ~5 y of age.
  • Fasting is not required for Lp(a) measurement, and despite being genetically determined, levels may be influenced in the presence of inflammation.
  • Because Lp(a) is genetically transmitted, youth whose parents have an elevated Lp(a) level are reasonable candidates for screening (forward cascade); conversely, reverse cascade screening is recommended when a child is found to have an elevated level of Lp(a).
  • Even if the absence of approved Lp(a)-lowering medications in youth, it is important to emphasize early and lifelong adoption of a heart-healthy lifestyle by the child and family members, especially with respect to smoking avoidance or cessation, given the thrombotic risk attributable to Lp(a).
  • Measurement of Lp(a) in youth with a history of ischemic stroke may be reasonable.

Specific Treatment goals

  • In statin-treated patients, a high Lp(a) is an independent ASCVD risk factor.
  • In primary prevention for adults aged 40–75 with 10-y ASCVD risk of 7.5–20% (intermediate risk), an Lp(a) ≥50 mg/dL or ≥100 nmol/L is reasonable to be used as a risk-enhancing factor to favor initiation of a moderate or high-intensity statin.
  • In high-risk or very-high-risk patients with LDLc ≥70 mg/dL (nonHDLc ≥100 mg/dL) + Lp(a) ≥50 mg/dL or ≥100 nmol/L on maximally tolerated statin intensity:

    • It is reasonable to consider more intensive therapies (such as ezetimibe and/or PCSK9 inhibitors) to lower LDLc (and nonHDLc) to achieve greater ASCVD risk reduction. 
  • In patients with very-high-risk ASCVD + baseline LDLc ≥70 mg/dL or nonHDLc ≥100 mg/dL despite maximally tolerated statin ± ezetimibe

    • The presence of an elevated Lp(a) may be used as a factor favoring addition of a PCSK9 inhibitor.
  • Although Niacin and HRT can reduce Lp(a) levels, these drugs are not recommended because of no demonstrated ASCVD benefit and the possibility of harm.