|Values are valid only on day of printing.|
Low-density lipoprotein cholesterol (LDL-C) is widely recognized as an established cardiovascular risk marker predicated on results from numerous clinical trials that demonstrate the ability of LDL-C to independently predict development and progression of coronary heart disease. In the United States, LDL-C remains the primary focus for cardiovascular risk assessment and evaluation of pharmacologic effectiveness based on treatment target goals. Considerable educational efforts have been invested and directed towards physicians, laboratorians, allied health staff, and the general public regarding LDL-C and strategies to lower it to reduce cardiovascular risk. Yet, a large body of evidence indicates that a narrow focus on LDL-C assessment and treatment alone is not the optimal strategy for patient care.
Several known limitations make LDL-C a less accurate marker of cardiovascular risk than either non-high-density lipoprotein cholesterol (non-HDL-C), LDL particle number, or apolipoprotein B (apoB). Furthermore, other triglyceride-rich lipoproteins also are atherogenic, including very-low-density lipoprotein (VLDL) remnants and intermediate-density lipoproteins (IDL). There continue to be numerous patients who succeed in meeting their target “LDL-C goal” but still develop complications from atherosclerotic vascular disease and suffer from cardiovascular events. These patients bear the burden of having residual risk not identified using traditional metabolic and cardiovascular markers. Here we discuss the role of alternate measurements of atherogenic particle concentrations beyond LDL-C, including non-HDL-C, LDL particle number, and apoB.
Low-Density Lipoprotein Cholesterol (LDL-C)
Low-density lipoproteins are a heterogeneous population of lipid particles classically defined as having a density of 1.006 to 1.063 kg/L obtained by preparative ultracentrifugation.1 The gold standard β-quantification method combines ultracentrifugation with precipitation and yields a collective quantitative measurement of LDL-C, intermediate-density lipoprotein cholesterol (IDL-C), and lipoprotein (a) cholesterol (Lp[a]-C). In practice, LDL-C is most commonly reported using the Friedewald equation (discussed below), which yields a fair estimation of LDL-C compared to β-quantification.2
A standard fasting lipid profile includes direct measurement of total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), and triglycerides (TG). The Friedewald equation calculates LDL-C by subtracting the HDL-C and VLDL-C (or TG/5) from the total cholesterol (LDL-C = TC-HDL-TG/5).3 The shortcomings of the Friedewald calculated LDL-C are addressed elsewhere.1,4 Importantly, there are limitations in the accuracy that require multiple fasting samples to be tested prior to initiation or modification of therapy and recommendations against reporting a calculated LDL-C in patients who are nonfasting, have triglycerides greater than 400 mg/dL, or have type III hyperlipoproteinemia. The equation is particularly inaccurate once the triglycerides are above 200 mg/dL or at low LDL-C concentrations. In particular, at lower LDL-C concentrations near the cutoff of 100 mg/dL, there is an error of plus or minus 15 mg/dL, indicating the “true” LDL cholesterol is somewhere between 85 and 115 mg/dL.5 This presents a major opportunity for misclassification of patients in terms of risk assessment and management. Thus, calculated LDL provides only marginal reflection of true LDL cholesterol concentration.
Homogeneous or “direct” LDL-C assays were designed to circumvent the issues with calculating LDL-C in specimens with triglycerides over 400 mg/dL. Clinical laboratories routinely utilize direct LDL-C assays for that purpose, however recent studies clearly indicate these methods, like lower calculated LDL-C, are unable to meet the National Cholesterol Education Program (NCEP) total error goal of <12% for LDL-C and are particularly unsuitable for use in a dyslipidemic population.6
Non-High-Density Lipoprotein Cholesterol (non-HDL-C)
Non-high-density lipoprotein cholesterol (non-HDL-C) is simply the difference between the total cholesterol concentration and the HDL cholesterol concentration, providing an estimate of cholesterol in the atherogenic particles including IDL, VLDL, Lp(a), and LDL. Non-HDL-C is not a novel concept and was recommended as a secondary target by the National Cholesterol Education Program Adult Treatment Panel III (NCEP ATP III) guidelines in 2001 for patients with triglycerides greater than 200 mg/dL.5 Advocacy for non-HDL-C began following widespread recognition of its superiority over LDL-C as a measurement of vascular event risk and demonstrated equivalency to apoB or LDL particle number in some clinical trials.7
In contrast to the standard fasting lipid profile, non-HDL-C may be calculated on nonfasting specimens and may avoid the problem of calculating LDL-C with high triglycerides, essentially making the need for a direct LDL-C assay obsolete. Furthermore, in patients with severe hypertriglyceridemia (>500 mg/dL), the current American Heart Association (AHA) guidelines recommend primary treatment of triglycerides to less than 500 mg/dL; only then is LDL-C targeted for secondary treatment.8 If a general assessment of atherogenic particles is desired in this population, either non-HDL-C, apoB, or LDL particle number can be measured to gain insight into the pathophysiology behind the disorder. While non-HDL-C may still be calculated when triglycerides are high, it is worth noting that analytical issues in homogeneous HDL-C assays require laboratories to use a conservative cutoff for assessment of lipemia.
NCEP-recommended cutpoints for non-HDL-C were arbitrarily set 30 mg/dL above the LDL-C cutpoints and assume triglyceride levels of 150 mg/dL or lower, and therefore a calculated VLDL-C of 30 mg/dL or less (ie, a triglyceride of 150 mg/dL divided by 5). The non-HDL-C goal is less than 100 mg/dL for patients who meet 1 of the following criteria: (1) established cardiovascular disease plus diabetes, (2) established cardiovascular disease plus multiple poorly controlled risk factors, (3) multiple risk factors of the metabolic syndrome, or (4) patients with acute coronary syndrome and/or coronary heart disease (Table 1). Guidelines from multiple societies (American Diabetes Association [ADA], American College of Cardiology [ACC], and National Lipid Association) recommend reporting of non-HDL-C on all lipid profiles, as it incurs no additional expense to the patient.7,9 Despite the length of time since the NCEP ATP III recommendations, some laboratories and physicians have been slow to adopt reporting and using non-HDL-C in clinical practice, perhaps due to uncertainty surrounding the meaning of what non-HDL-C truly represents and/or due to the amount of education already vested in targeting LDL-C with statin therapies.
|Very high riska||<100 (optional <70)||<130 (optional <100)|
|High risk: CHDb or CHD risk equivalentc||<100||<130|
|Moderately high risk: ≥2 risk factorsd (10-y risk, 10% to 20%)||<130 (optional <100)||<160 (optional <130)|
|Moderate risk: ≥2 risk factorsd (10-y risk <10%)||<130||<160|
Table 1. ATP III LDL-C and Non-HDL-C Goals and Cutpoints for Drug Therapy
a Very high risk is defined as established CVD plus diabetes, or plus multiple poorly controlled risk factors, or multiple risk factors of the metabolic syndrome (especially triglycerides >200 mg/dL), or patients with acute coronary syndrome
b CHD includes history of myocardial infarction, unstable angina, coronary artery procedures (bypass or angioplasty), or evidence of clinically important myocardial ischemia
c CHD risk equivalents are defined as clinical manifestations of noncoronary atherosclerotic disease, including peripheral arterial disease, abdominal aortic aneurysm, and carotid artery disease, diabetes and ≥2 risk factors with 10-y risk of CHD >20%
d Risk factors include cigarette smoking, hypertension (blood pressure >140/90 mm Hg or antihypertensive medication), low HDL-C (<40 mg/dL), family history of CHD in first-degree relative, or age (>45 y in men, >55 y in women) Abbreviations: ATP III; adult treatment panel III; CVD, cardiovascular disease; CHD, coronary heart disease
The benefit of non-HDL-C may arguably be cost, as it can be calculated from a standard lipid panel without additional expense. It also is unquestionably superior to LDL-C for prediction of risk. However, the compromised accuracy of HDL-C assays in patients with hypertriglyceridemia reduces the benefit in reporting non-HDL-C. Nevertheless, the impediment to widespread implementation and utilization of non-HDL-C is primarily educational. It is often not intuitive for physicians and laboratorians to understand, interpret, or make decisions upon a non-HDL-C concentration.
Apolipoprotein B is the structural protein for all of the atherogenic lipoproteins (VLDL-C, IDL-C, LDL-C) and modulates the transportation of lipids from the liver and gut to peripheral tissues. There are 2 isoforms of apoB: apoB-100, which is the major isoform and is synthesized in the hepatocytes, and apoB-48, which is synthesized in the intestines and is the structural protein of chylomicrons. Each atherogenic particle contains just 1 molecule of apoB, a critical concept that allows apoB to directly translate into an assessment of atherogenic particle number. Approximately 90% of apoB particles are LDL particles, likely due to the longer half-life compared to VLDL particles (3 to 4 days versus 3 to 4 hours, respectively). Thus, given their smaller size and greater length of time in circulation, the number of LDL particles, opposed to the number of VLDL particles, determines cardiovascular risk.
The process by which atherosclerotic lesions develop is unquestionably initiated once an apoB particle becomes trapped within the vascular wall.10 The apoB particle is degraded and the cholesterol contained within the particle becomes retained by macrophages. This sequence of events drives an additional cascade of inflammatory processes and eventual formation of complex atherosclerotic lesions, ultimately manifesting in clinical events. Therefore, apolipoprotein B is not simply a measurement of risk, like LDL-C or non-HDL-C, but is actually causative in the progression of atherosclerosis.
Standardization of apoB assays has largely been successful due to availability of suitable reference materials and support from the International Federation of Clinical Chemistry (IFCC).11 ApoB measurements can be performed with a higher degree of accuracy compared to assays for HDL-C, triglycerides, or LDL-C despite widespread belief that these assays are currently standardized. Between-laboratory assessments have repeatedly demonstrated a coefficient of variation between 3.1% and 6.7%. Fasting is not necessary and assays are widely commercially available on a variety of automated platforms. Therefore, there are no analytical hindrances toward implementation of an apoB assay in the routine clinical laboratory.
A Consensus Conference Report in 2008 from the ADA and the ACC, along with an international expert panel, recommended that measurement of apoB be included along with LDL-C and non-HDL-C in a variety of high- and highest-risk patients (Table 2).9 The high-risk categories encompass patients with diabetes, known cardiovascular disease, and/or risk factors for cardiovascular disease. In 2009, the Canadian guidelines on management of dyslipidemia also recommended a goal apoB in routine lipid assessments (<80 mg/dL in high-risk patients).12
Table 2. ADA/ACC Treatment Goals in Patients with Lipoprotein
Abnormalities and Cardiometabolic Risk
a Highest-risk patients include those with either known CVD or diabetes plus ≥1 additional major CVD risk factor
b High-risk patients include those with no diabetes or known CVD but ≥2 major additional risk factors or diabetes but no other major risk factors*
* Other major risk factors beyond dyslipoproteinemia include smoking, hypertension, and family history of premature coronary artery disease
Abbreviations: ADA, American Diabetes Association; ACC, American College of Cardiology; CVD, cardiovascular disease
Low-density lipoprotein particle number
Measurement of LDL particle concentration by nuclear magnetic resonance (NMR) spectroscopy, also referred to as LDL particle number, is a novel methodology that provides quantitation of size and concentration of LDL particles and various lipid subfractions including VLDL, IDL, LDL, and HDL. Lipoproteins are analyzed by the NMR according to the spectral signals produced by the terminal methyl groups contained within the lipid particles. The number of methyl groups present on triglycerides, cholesterol, and phospholipids is consistent for particles at a given size, allowing for translation into particle concentration. NMR lipoprotein analyses have been evaluated against other existing methods of lipoprotein subclass methods, however, currently there are no standardization programs targeted towards these analytes.13 Regardless, both primary and secondary prevention trials have demonstrated that LDL particle number is superior to LDL-C, consistent with analyses conducted with apoB. While early studies emphasized the atherogenicity of small LDL particles, we now know that all LDL particles are atherogenic, as evidenced by patients with familial hypercholesterolemia who have large, buoyant LDL particles and early atherosclerosis. Thus, the primary focus should remain on reduction of the number (ie, concentration) of LDL particles, without a significant amount of effort to distinguish between large and/or small LDL particles.
Multiple epidemiological and clinical trials support apoB and LDL particle number as the superior marker for cardiovascular risk prediction when compared to non-HDL-C and LDL-C (Table 3). The INTERHEART study was a large international standardized case-control study of acute myocardial infarction involving more than 12,000 subjects, over 14,000 age- and sex-matched controls, and a variety of ethnic groups.14 ApoB had the highest odds ratio for prediction of vascular disease and emerged as the superior marker over non-HDL-C across all ethnic groups. Apolipoprotein-Related Mortality Risk (AMORIS) was another large epidemiologic study, with over 175,000 asymptomatic subjects.15 After adjusting for age and traditional lipid risk factors, apoB remained a significant predictor of myocardial infarction in both men and women while LDL-C was no longer a risk factor in females and only modestly predictive of myocardial infarction in males. The AMORIS trial also demonstrated that apoB was a better predictor of risk than LDL-C, even in individuals with concentrations of LDL-C below the median. This was a significant finding because up to 50% of patients with coronary heart disease will have a normal cholesterol (<200 mg/dL), yet a large proportion will have elevated apoB concentrations and therefore, residual risk.
|The City Heart Study||9,231||P||M/F||A||Equivalent|
|Women’s Heath Study||15,362||P||F||A||Equivalent|
|Nurse’s Health Study||234/468||CC||F||A||Equivalent|
|Health Professionals’ Follow-up Study||103/643||NCC||M||DM||Equivalent|
|Bogalusa Heart Study||1,061||P||M/F||A||Equivalent|
|Leiden Heart Study||848||CT||M/F||CAD||apoB|
|Casale Monferrato Study||1,565||P||M/F||DM2||apoB|
|Health Professionals Follow-up Study||266/532||NCC||M||A||apoB|
|The Chinese Heart Study||3,586||P||M/F||A||apoB|
|Framingham Offspring Study||3,066||P||M/F||A||LDL particles|
|Cardiovascular Risk in Young Finns Study||879||P||M/F||A||apoB|
Table 3. Epidemiologic and Clinical Trial Studies Compare Predictive Markers of Cardiovascular Risk
Abbreviations: P, prospective; CC, case control; NCC, nested case control; CT, clinical trial; XS, cross sectional; A, asymptomatic; DM, diabetes mellitus; CAD, coronary artery disease; MI, myocardial infarction
There is controversy surrounding the potential redundancy of non-HDL-C, apoB, and LDL particle number because, in theory, they all yield an assessment or estimate of atherogenic particle number. In large population-based studies they will provide equivalent predictive value. However, despite high overall correlation, for any number of individuals there will be a large amount of discordance among the 3 parameters. The greatest discordance is noted in patients with hypertriglyceridemia or other dyslipoproteinemias (familial combined hyperlipidemia, familial dysbetalipoproteinemia, familial hypoalphalipoproteinemia), and none of those dyslipoproteinemias can be recognized with non-HDL-C. Importantly, non-HDL-C is only a collective assessment of the cholesterol content in lipoprotein particles and provides no information regarding atherogenic particle number, potentially leaving a potential large number of patients with residual risk that remains even on treatment therapies.
Several large clinical trials have demonstrated superiority in utilization of apoB over LDL-C or non-HDL-C for assessment of residual risk in patients receiving lipid-lowering therapies, particularly as LDL-C target goals becoming increasingly more aggressive. In the Air Force/Texas Coronary Atherosclerosis Prevention Study (AFCAPS/TexCAPS) trial, 6,600 participants received lovastatin for 1 year, which resulted in a 37% reduction in the risk of an acute coronary event.16 At the onset of the trial, the association between LDL-C, apoB, and the risk of a major coronary event was similar. However, after 1 year of statin therapy only, on-treatment apoB emerged as a strong risk predictor (p=0.01) compared to LDL-C, which lost statistical significance (p=0.162).
Similarly, data combined from 2 randomized secondary prevention trials, Treating to New Targets (TNT) and Incremental Decrease in End Points through Aggressive Lipid Lowering (IDEAL), demonstrate additional opportunities for more aggressive statin therapy when using apoB rather than non-HDL-C.17 The 2 trials showed that non-HDL-C and apoB had equal predictive power in identifying the risk of a future clinical event but the 2 markers differed in their ability to judge the adequacy of statin therapy. However, this can be explained by the fact that different percentiles were used for apoB and non-HDL-C concentrations. A significant number of patients on high-dose (80 mg) atorvastatin were identified as having undesirable levels of apoB (reduced to the 30th percentile) yet succeeded in achieving both non-HDL-C and LDL-C goals (reduced to the 5th percentile). It is currently unknown if altering treatment strategies to achieve an apoB goal would demonstrate an improvement in outcomes.
Statin therapies, either alone or in combination with ezetimibe, a cholesterol-absorption inhibitor, will lower LDL-C, non-HDL-C, and apoB/LDL particle number but not to the same degree (Table 4).18 Various hypotheses exist for why apoB is superior to LDL-C for determination of residual risk in patients on statin therapy. Statins are designed to inhibit 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase and restrict production of cholesterol. Therefore, use of statins alters the cholesterol content of lipoproteins to a greater extent than the actual particle number, resulting in a greater number of patients who will reach their LDL-C or non-HDL-C “goals” but will unknowingly continue to have an elevated number of atherogenic particles.
with therapy (%)
|Mean concentration achieved
with therapy (mg/dL)
|Mean population percentile
achieved with therapy (%)
Table 4. Effect of Statin Therapy on LDL-C, Non-HDL-C, and ApoB
The ADA/ACC Consensus Conference Report and the Canadian Cardiovascular Society have recommended apoB goals for treatment of dyslipidemia and prevention of cardiovascular disease, as outlined in Table 2. The ADA/ACC also recommends an apoB goal of 80 mg/dL in addition to an LDL-C goal of 70 mg/dL and a non-HDL-C goal of 100 mg/dL for patients with either established cardiovascular disease or diabetes with 1 risk factor. In patients without cardiovascular disease but 2 risk factors present, an apoB goal of 90 mg/dL is recommended by the ADA/ACC.9
Clinical trials lend support to the apoB goals recommended in the guidelines. The Pravastatin or Atorvastatin Evaluation and Infection Therapy (PROVE IT) trial evaluated more than 4,000 subjects who were hospitalized for acute coronary syndrome with a baseline median apoB concentration near 100 mg/dL.19 Patients were randomized to receive treatment with either 40 mg of pravastatin (moderate therapy) or 80 mg of atorvastatin (intensive therapy). At conclusion of the trial, the median apoB level for the moderate and intensive therapy groups was 90 mg/dL and 67 mg/dL, respectively. The patients in the intensive therapy group experienced a 16% reduction in the hazard ratio for death or major cardiovascular event (p=0.005, 95% CI: 5%-26%) compared to patients in the moderate therapy group. In the Collaborative Atorvastatin Diabetes Study (CARDS), more than 2,800 diabetic subjects with no history of cardiovascular disease were randomized to receive either low-dose atorvastatin (10 mg) or placebo.20 The mean baseline concentration of LDL-C, non-HDL-C, and apoB in the treatment group was 118 mg/dL, 153 mg/dL, and 117 mg/dL, respectively. At 1 year, the atorvastatin group had a significant decrease in LDL-C (mean: 72 mg/dL, 40.9% reduction) and less reduction in either non-HDL-C (mean: 101 mg/dL, 38.1% reduction) or apoB (mean: 80 mg/dL, 24.3% reduction) compared to the placebo group (mean LDL-C, non-HDL-C, and apoB: 120 mg/dL, 157 mg/dL, and 110 mg/dL, respectively). The treatment arm had a 37% risk reduction for major cardiovascular events, leading to initiation of statin therapy in diabetics becoming routine in clinical practice. It is unknown if further risk reduction could be seen if apoB is used as the primary treatment goal in that population.
In the Justification for the Use of Statins in Primary Prevention: An Intervention Trial Evaluating Rosuvastatin (JUPITER) trial, patients all had low LDL-C (mean 108 mg/dL, Framingham 25th percentile) and low non-HDL-C (mean 134 mg/dL, Framingham 30th percentile) yet benefited greatly from statin therapy because of an elevated C-reactive protein (>4.0 mg/dL). Interestingly, JUPITER patients also had a moderately high apoB (mean 109 mg/dL, Framingham 60th percentile for males and 70th percentile for females) and therefore it is plausible apoB may have just as effectively identified patients for statin therapy and the outcome of that trial should not be surprising when one accounts for the discordance among LDL-C, non-HDL-C, and apoB.
ApoB also helps identify certain dyslipidemias like familial combined hyperlipidemia (FCH), which is the most common familial atherogenic dyslipoproteinemia and is present in 25% to 40% of patients who present with premature myocardial infarction. The FCH phenotype can be identified with a standard algorithm utilizing total cholesterol, triglycerides, and apoB. Non-HDL-C cannot substitute for apoB in the diagnosis of FCH and a few other dyslipoproteinemias.21
ApoB likely plays an important role in development of vascular disease and ample and compelling evidence exists to support routine utilization and integration of apoB into practice. Defined apoB treatment goals are now embedded in clinical practice guidelines. Treatment to an apoB goal, as opposed to LDL-C or non-HDL-C, provides significant opportunity to improve patient care and cardiovascular risk stratification in moderate- to high-risk groups. Furthermore, apoB can identify dislipidemias that non-HDL-C would miss. Laboratories should consider adding apoB to their test menu and clinicians should begin to evaluate this parameter. Given available clinical information, and its lack of added cost, laboratories should, at a minimum, report non-HDL-C with all lipid panels and include clear interpretive guidelines for physicians and other allied health staff. Further educational efforts will need to be directed toward utilization and integration of non-HDL-C and/or apoB into clinical practice.
Authored by: Dr. Amy Saenger