Journal of the American College of Cardiology Vol. 50, No.

18, 2007
© 2007 by the American College of Cardiology Foundation ISSN 0735-1097/07/$32.00
Published by Elsevier Inc. doi:10.1016/j.jacc.2007.07.045

STATE-OF-THE-ART PAPER

Beyond Low-Density Lipoprotein Cholesterol

Defining the Role of Low-Density
Lipoprotein Heterogeneity in Coronary Artery Disease
James O. Mudd, MD,* Barry A. Borlaug, MD,* Peter V. Johnston, MD,* Brian G. Kral, MD, MPH,*
Rosanne Rouf, MD,* Roger S. Blumenthal, MD,* Peter O. Kwiterovich, JR, MD†
Baltimore, Maryland

Recent clinical trials in patients with coronary artery disease (CAD) provide evidence that low-density lipoprotein
cholesterol (LDL-C) levels should be lowered even further to prevent recurrent CAD. However, despite more ag-
gressive interventions for lowering LDL-C levels, the majority of CAD events go undeterred, perhaps related to
the fact that intervention was not started earlier in life or that LDL-C levels represent an incomplete picture of
atherogenic potential. Nevertheless, LDL-C remains the contemporary standard as the primary goal for aggres-
sive LDL reduction. If triglycerides are ⬎200 mg/dl, the measurement of non–high-density lipoprotein choles-
terol (HDL-C) is recommended. Measurement of apolipoprotein (apo)B has been shown in nearly all studies to
outperform LDL-C and non–HDL-C as a predictor of CAD events and as an index of residual CAD risk. This is be-
cause apoB reflects the total number of atherogenic apoB-containing lipoproteins and is a superior predictor of
the number of low-density lipoprotein particles (LDL-P). Estimates of LDL-P and size can also be made by nu-
clear magnetic resonance spectroscopy, density gradient ultracentrifugation, and gradient gel electrophoresis.
Although a number of studies show that such estimates predict CAD, LDL-P, and size often accompany low
HDL-C and high triglyceride levels, and therefore such additional lipoprotein testing has not been recommended
for routine screening and follow-up. Because apoB is a superior predictor of LDL-P, we recommend that apoB
and the apoB/apoA-I ratio be determined after measurement of LDL-C, non–HDL-C, and the ratio of total
cholesterol/HDL-C to better predict CAD and assess efficacy of treatment. (J Am Coll Cardiol 2007;50:
1735–41) © 2007 by the American College of Cardiology Foundation

Elevated low-density lipoprotein cholesterol (LDL-C) is a
well-established independent risk factor for coronary artery
disease (CAD). A number of primary and secondary trials
have demonstrated that lowering of LDL-C decreases the
incidence of CAD. These trials have been reviewed in detail
elsewhere (1– 4) and include treatment with a fat-modified
diet and therapeutic agents such as inhibitors of 3-hydroxy-
3-methylglutaryl coenzyme A reductase (statins), fibrates,
niacin, and bile acid sequestrants. These studies, in aggre-
gate, are the basis for the recommendations of the National
Cholesterol Education Program (NCEP) that LDL-C be
lowered to ⬍100 mg/dl in patients with CAD or CAD risk
equivalents, such as diabetes or peripheral arterial disease.
The HPS (Heart Protection Study) (5), the PROVE-IT
(Pravastatin or Atorvastatin Evaluation and Infection Ther-
apy) study (6), and the REVERSAL (Reversal of Athero-
sclerosis with Aggressive Lipid Lowering) study (7) all
indicated a significant benefit to lowering LDL-C well
below 100 mg/dl. The NCEP-Adult Treatment Panel
From the *Johns Hopkins Ciccarone Preventive Cardiology Center and the †Johns
Hopkins University Lipid Clinic, Baltimore, Maryland.
Manuscript received April 30, 2007; revised manuscript received July 16, 2007,
accepted July 17, 2007.
(ATP) III guidelines were therefore amended in 2004 by
adding an optional goal of LDL-C ⬍70 mg/dl in patients
with recent acute coronary syndromes or those otherwise
felt to be at “very high risk” for adverse cardiovascular events
(2). Subsequently, the TNT (Treating to New Targets)
study (8), with atorvastatin 80 mg/day versus 10 mg/day,
and ASTEROID (A Study to Evaluate the Effect of
Rosuvastatin [40 mg/day] on Intravascular Ultrasound-
Derived Coronary Atheroma Burden) (9) both supported
the more aggressive treatment of LDL-C to ⬍70 mg/dl.
O’Keefe et al. (3) proposed that the threshold for atheroscle-
rotic progression may be at an LDL-C level of ⬍70 mg/dl.
Still, despite aggressive use of statins, the majority of
CAD events are not prevented (1– 8). In the IDEAL
(Incremental Decrease in End Points Through Aggressive
Lipid Lowering) study, the use of atorvastatin 80 mg/day
compared with simvastatin 20 mg/day did not result in a
significant reduction in the primary outcome of major
coronary events (10). This may be due to the fact that these
studies were of relatively short duration and in older adults.
The benefit is likely to be greater if intervention is started
earlier in life and continued for a longer time period (11).
The presence of traditional risk factors such as hypertension,
diabetes, obesity, elevated triglycerides (TG), and low high-
1736 Mudd et al. JACC Vol. 50, No. 18, 2007
LDL Heterogeneity in CAD October 30, 2007:1735–41

Abbreviations density lipoprotein cholesterol Biochemical Basis of LDL Heterogeneity

and Acronyms (HDL-C) may also render the
LDL-C intervention less effec- Low-density lipoprotein is a heterogeneous group of parti-
apo ⴝ apolipoprotein cles that vary in their core content of cholesterol (12–14). A
tive. The influence of nontradi-
CAD ⴝ coronary artery greater amount of cholesterol in LDL creates larger, more
disease
tional risk factors such as small,
dense LDL particles, elevated lev- buoyant particles (sometimes referred to as LDL subclass A).
CE ⴝ cholesteryl esters A lower amount of cholesterol in LDL generates smaller,
els of high-sensitivity C-reactive
DGU ⴝ density-gradient denser particles (sometimes referred to as LDL subclass B).
ultracentrifugation
protein (hs-CRP), lipoprotein
(a) [Lp(a)], homocysteine, and Each LDL particle, regardless of its cholesterol content, has
GGE ⴝ gradient gel
unknown genetic and environ- 1 molecule of apoB for each LDL molecule (13). For 2
electrophoresis
mental risk factors on recurrent patients with the same LDL-C, the one with a preponder-
HDL-C ⴝ high-density
lipoprotein cholesterol CAD events is not completely ance of small LDL particles will have a greater LDL-P,
understood (1,2). carry a significantly greater risk of CAD, and may therefore
LDL-C ⴝ low-density
lipoprotein cholesterol Another possible contributory benefit from more aggressive therapy (12–16).
LDL-P ⴝ total number of
factor is that LDL-C does not VLDL–IDL–LDL pathway. Alterations in the very-low-
low-density lipoprotein reflect the atherogenicity of all density lipoprotein (VLDL)–intermediate-density lipoprotein
particles of the apolipoprotein (apo)B- (IDL)–LDL pathway, whether owing to genetic mutations,
Lp(a) ⴝ lipoprotein (a) containing lipoproteins nor does acquired disorders, diet, or other factors, result in variation in
NMR ⴝ nuclear magnetic it necessarily represent the total particle composition, size, and abundance (12–16) (Fig. 1). For
resonance number of low-density lipopro- example, the metabolic syndrome, diabetes, familial combined
TC ⴝ total cholesterol tein particles (LDL-P) or the hyperlipidemia, and hyperapobetalipoproteinemia (hyper-
TG ⴝ triglycerides distribution of size within those apoB) are conditions in which increased synthesis and secre-
particles. This is particularly true tion of VLDL leads to the overproduction of small, dense
if small, dense LDL-P is ele- LDL and HDL particles, a combination that increases cardio-
vated, leading to an underestimate of LDL-P by LDL-C. vascular risk (13–17) (Fig. 1). Insulin resistance or a defect in
Low HDL-C with a normal LDL-C is often accompanied the normal activity of the acylation stimulatory protein can
by increased small, dense LDL-P (12). Patients with cause higher plasma free fatty acid levels, leading to greater
hyper-TG may also have increased small, dense LDL-P, hepatic free fatty acid uptake, increased TG production,
placing them at higher CAD risk than those hyper-TG decreased hydrolysis of apoB, and increased production and
patients with a normal LDL-P (13). Here we review the secretion of larger TG-rich VLDL particles (13–17) (Fig. 1).
epidemiologic and clinical evidence that determination of An increased exchange of TG from VLDL for cholesteryl
LDL-P supplements the LDL-C measurement and has ester (CE) in LDL and HDL occurs via the CE transfer
both diagnostic and therapeutic implications. protein, resulting in cholesterol-depleted TG-enriched li-

Figure 1 Mechanisms of the Production of the “Dyslipidemic Triad”

An increased flux of free fatty acids (FFA) from adipose tissue can result from insulin resistance or a defect in the acylation stimulatory protein (ASP) pathway. Enhanced
hepatic uptake of FFA leads to increased triglyceride (TG), apolipoprotein (apo) B, and very-low-density lipoprotein (VLDL) production. The TG in VLDL is exchanged for
cholesteryl esters (CE) from low-density lipoproteins (LDL) and high-density lipoproteins (HDL) by the cholesteryl ester transport protein (CETP), producing CE-depleted
LDL and HDL. The TG in the core of LDL and HDL is then hydrolyzed by lipases, producing both small, dense LDL and HDL. Such HDL is more likely to be excreted by
the kidney, resulting in low HDL-C levels. HL ⫽ hepatic lipase; IDL ⫽ intermediate-density lipoprotein; IR ⫽ insulin resistance; LCAT ⫽ lecithin-cholesterol acyltrans-
ferase; LPL ⫽ lipoprotein lipase.
JACC Vol. 50, No. 18, 2007 Mudd et al. 1737
October 30, 2007:1735–41 LDL Heterogeneity in CAD

poproteins that are subsequently converted into smaller, denser large, prospective studies such as AMORIS (Apoprotein-
particles by hepatic lipase (Fig. 1). The TG-enriched LDL is Related Mortality Risk Study) (29), the Health Professionals
hydrolyzed, leading to increased small, dense LDL-P (15). Follow-Up Study (27), the Quebec Cardiovascular Study
The TG-rich HDL is also modified by hepatic lipase, produc- (25,26), and the study by Moss et al. (30).
ing smaller HDL that is cleared more rapidly by the kidneys, Benn et al. (31) recently reported that apoB had a higher
contributing to lower levels of HDL-C and its major predictive ability than LDL-C for ischemic cardiovascular
apolipoprotein, apoA-I (Fig. 1). Such abnormalities in the disease in 9,231 asymptomatic Danish men and women.
VLDL–IDL–LDL pathway are often manifested by the Using receiver-operating characteristic curves, they found
dyslipidemic triad of hyper-TG, increased small, dense that the predictive ability of non–HDL-C was similar to
LDL-P, and low HDL-C. apoB. However, the interindividual biological variation for
apoB (coefficient of variation [CV] 6.9%) was superior to
The Contemporary Standards: non–HDL-C (TC CV 6% plus HDL-C CV 7.1%), indi-
LDL-C and Non–HDL-C cating that repeated measurements of apoB are more reliable
in a single patient. Apolipoprotein B also appears to be a
LDL-C. Aggressive reduction of LDL-C is the current
better predictor of subsequent CAD events in patients on
primary goal of lipid-lowering therapy (1–10). The LDL-C
treatment with statins (28,32).
reflects the cholesterol content of several atherogenic li-
Ratio of apoB to apoA-I. Results from a number of
poproteins, including LDL, IDL, and Lp(a). The conven-
epidemiologic studies indicated that the ratio of apoB to
tional lipid panel measures levels of total cholesterol (TC),
apoA-I was the strongest predictor of CAD (28). In both
HDL-C, and TG and calculates LDL-C using the Friede-
the placebo and the treatment groups from the AFCAPS/
wald equation (18). To use this formula, the patient must be
TEXCAPS (Air Force/Texas Coronary Atherosclerosis
fasting with a TG level ⬍400 mg/dl. As LDL-C nears goal,
Prevention Study) trial, the apoB/apoA-I ratio was a robust
the inaccuracies of the Friedewald equation are amplified
predictor of CAD (32). Similarly, van Lennep et al. (33)
(19). For example, calculated LDL-C generally underesti-
reported in patients with CAD on statin therapy that only
mates the true concentration of small, dense LDL particles,
apoB and the apoB/apoA-I ratio predicted myocardial
especially when the TG level is ⬎200 mg/dl (13). Given these
infarction and all-cause mortality, whereas LDL-C and TG
limitations of LDL-C, the NCEP recommended using
did not. In both men and women from the AMORIS
non–HDL-C (total cholesterol ⫺ HDL-C) as a secondary
(Apolipoprotein-Related Mortality Risk) study, the superi-
target of therapy when the TG level is ⬎200 mg/dl (1).
ority of the ratio of apoB to apoA-I, compared with the
Non–HDL-C. The non–HDL-C estimates the choles-
TC/HDL-C ratio, became more obvious as risk of CAD
terol concentration of all the apoB-containing lipoproteins,
increased (28). In men, the relative risk of CAD increased at
namely VLDL, IDL, LDL, and Lp(a), in contrast to
an apoB/apoA-I ratio of 1.0, and the comparative figure for
LDL-C, which does not include VLDL-C. The non–
females was 0.86 (28). These results in aggregate are not
HDL-C can be measured nonfasting. If the TG level is
surprising, given that increased LDL-P is usually accompa-
elevated to ⬎200 mg/dl, the non–HDL level will reflect the
nied by low HDL and apoA-I (Fig. 1).
increase in VLDL-C. The non–HDL-C appears superior
The percentile distributions for LDL-C, non–HDL-C,
to LDL-C estimation in establishing CAD risk and monitor-
and apoB are available from the Framingham Heart Study
ing treatment (20 –23). However, lipid levels do not necessarily
(23,34,35) (Table 1). For a particular goal for LDL-C
equate to lipoprotein particle levels, and non–HDL-C does
treatment from the NCEP-ATP III (e.g., ⬍100 mg/dl) one
not permit a determination of whether there is also an
can estimate what the equivalent goal (based on percent
increase in small, dense LDL-P in the hyper-TG patient.
in the population) for apoB might be (e.g., ⬍80 mg/dl)
(Table 1). The percentile distributions for HDL-C and
Beyond LDL-C and Non–HDL-C
Estimated
LDL-C, Non–HDL-C,
Percentiles
ApoB,
of and LDL-P
ApoB measurement. The apoB level reflects the total Table 1
Estimated Percentiles of
number of atherogenic apoB-containing lipoproteins, be- LDL-C, Non–HDL-C, ApoB, and LDL-P
cause each chylomicron, chylomicron remnant, VLDL, Percentile
IDL, LDL, and Lp(a) particle contains 1 molecule of apoB.
5th 10th 25th 50th 75th 90th 95th
However, 90% of total plasma apoB is contained within the
LDL-C (mg/dl) 80 90 110 130 160 180 200
LDL particles (13). Thus, for a given LDL-C level, a higher Non–HDL-C 110 120 140 160 190 210 230
apoB level indicates higher LDL-P. Measurement of apoB (as (mg/dl)
well as apoA-I) does not need to be done in the fasting state, ApoB (mg/dl) 60 70 80 100 120 140 150
has been standardized by the World Health Organization, LDL-P (nmol/l) 800 900 1,100 1,400 1,800 2,000 2,100
and is available in most large commercial laboratories (24). Estimates were derived from the Framingham Heart Study (23,34,36). Men and women were
Moreover, apoB has been shown in nearly all studies to combined.
ApoB ⫽ apolipoprotein B; LDL-C ⫽ low-density lipoprotein cholesterol; LDL-P ⫽ total number of
outperform LDL-C and non–HDL-C measurements in car- low-density lipoprotein particles; non–HDL-C ⫽ total cholesterol ⫺ high-density lipoprotein
diovascular risk stratification (25–30). These include several cholesterol.
1738 Mudd et al. JACC Vol. 50, No. 18, 2007
LDL Heterogeneity in CAD October 30, 2007:1735–41

Estimated
and ApoA-IPercentiles
Levels (mg/dl)
of HDL-C
in Men and Women
Estimated Percentiles of HDL-C
among elderly women. The LDL-P remained significant,
Table 2 even after adjustment for traditional risk factors, whereas
and ApoA-I Levels (mg/dl) in Men and Women
LDL particle size did not. Mackey et al. (40) found that
Percentile
LDL-P, small, dense LDL, and large VLDL were posi-
5th 10th 25th 50th 75th 90th 95th
tively associated with coronary artery calcification in healthy
HDL-C
postmenopausal women, after adjustment for age, systolic
Men 28 31 37 43 51 61 67
Women 35 39 46 55 66 77 84
blood pressure, current smoking, LDL-C, HDL-C, and
ApoA-I
TG. Blake et al. (15) found that LDL-P in healthy women
Men 92 99 114 130 153 178 196 was independently predictive of future myocardial infarc-
Women 107 116 132 154 181 206 224 tion, even after adjustment for the ratio of TC to LDL-C,
Estimates were derived from the Framingham Heart Study (35).
apoB, and hs-CRP.
ApoA-I ⫽ apolipoprotein A-I; HDL-C ⫽ high-density lipoprotein cholesterol. Rosenson et al. (41) measured LDL and HDL particle
size and concentration by NMR spectroscopy in frozen
apoA-I in men and women from the Framingham Heart
plasma samples of placebo- and pravastatin-treated subjects.
Study (35) are also provided in Table 2. We propose that in
Whereas lipid levels did not predict angiographic progres-
addition to apoB, the ratio of apoB/apoA-I be included as
alternative goals in consensus recommendations for LDL sion of CAD, LDL-P strongly correlated with progression
reduction and HDL increase. (41).
Other assessments of LDL particle number and size. In the MESA (Multiethnic Study of Atherosclerosis)
The LDL-P is determined by the rate at which the LDL study, Mora et al. (42) studied the association of LDL-P
particles are produced and cleared from the circulation and particle size by NMR with carotid intima-media
(13–16) (Figs. 1 and 2). Furthermore, there are documented thickness (IMT) in over 5,000 apparently healthy indi-
variations between different ages, genders, and races (36 – viduals. When controlling for traditional CAD risk
38). For a given LDL-C level, an increase in small, dense factors and LDL subclass correlation, both large and
LDL-P is associated with increased atherogenesis, which small LDL-P, but not LDL size, were significantly
may explain some of the variation in risk between different associated with carotid IMT.
individuals and populations (36 –38). More direct measures
of LDL-P may help to further delineate the risk of CAD
attributable to metabolic abnormalities such as diabetes and
the metabolic syndrome (14,16,17).
The direct measurement of LDL-P number and size is
technically difficult, relatively time consuming, and more
expensive than apoB measurement. The LDL-P number
and size have also been assessed by nuclear magnetic
resonance (NMR) spectroscopy, density-gradient ultracen-
trifugation (DGU), or gradient gel electrophoresis (GGE).
A recent report found that an assessment of these methods
to determine LDL size agreed poorly (39). The LDL-P
however, was not evaluated, but the different tests also
reported significant variations in LDL-C, which is what our
current treatment guidelines are based upon.
NMR spectroscopy. Nuclear magnetic resonance spectros-
copy rapidly determines the sizes and concentrations of 15
lipoprotein subclasses based on the spectral characteristics of
methyl groups within the lipid molecules and the difference
in size of the lipoprotein particles (12). The LDL-P by
NMR includes IDL and 3 LDL subclasses. Nuclear mag-
netic resonance has been validated against existing methods Proposed Biochemical and Cellular
Figure 2
Mechanisms of Enhanced Small, Dense LDL
of lipoprotein subclass determination (12); however, there is
no international standardization program. Nuclear magnetic Increased small dense low-density lipoprotein (LDL) production, combined with
a lower affinity of small, dense LDL for the low-density lipoprotein receptor
resonance was only performed by LipoScience (Raleigh,
(LDL-R) and a longer residence time in plasma, leads to an increased number
North Carolina). of small, dense LDL particles. The increased small, dense LDL cross the arte-
The LDL-P appears to be a particularly strong predictor rial wall more readily, bind more avidly to proteoglycans in the intimal matrix,
and are more easily oxidized, characteristics all conducive to atherogenesis
of CAD in women. In the prospective CHS (Cardiovascular
(16). Small, dense LDL promotes endothelial cell dysfunction, inducing greater
Health Study) (37), NMR-determined LDL-P and small production of plasminogen activator inhibitor (PAI)-I and thromboxane A2 (TXA2).
LDL particle size predicted incident CAD, primarily
JACC Vol. 50, No. 18, 2007
October 30, 2007:1735–41
In the prospective EPIC (European Prospective Investi-
gation Into Cancer and Nutrition)-Norfolk study (22) in
25,663 subjects, both LDL-P, as assessed by NMR, and
non–HDL-C were more closely associated with CAD than
LDL-C. After HDL-C and TG levels were accounted for,
LDL-P lost its discriminative power over LDL-C. Consis-
tent with the MESA study, the EPIC-Norfolk study also
found that LDL size was not predictive of CAD after
adjustment for LDL-P. These results argue against the
routine implementation of determining LDL-P number
and size by NMR for CAD risk assessment. However, the
fact that patients with low HDL-C and/or higher TG often
have elevated numbers of LDL-P, without having high
LDL-C, may enable their LDL-related CAD risk to be
managed more effectively. Although data from clinical
intervention studies (see subsequent text) support LDL-P
by NMR as an alternative treatment target (but not superior
to apoB), the value of LDL-P monitoring as an additional
treatment target beyond LDL-C needs to be assessed in
intervention trials.
DGU. In DGU, lipoprotein fractions from a density gra-
dient are eluted systematically and the cholesterol content
determined. The method allows an estimate of the relative
distribution of cholesterol but not particle number within
the IDL and LDL subclasses. Evidence from several angio-
graphic trials using DGU indicated that treatment benefit is
related to a significant decrease in small, dense LDL-C
(43– 45), independent of other lipid and nonlipid risk
factors.
The Vertical Auto Profile II (Atherotech, Birmingham,
Alabama) is a DGU-based test that determines the choles-
terol content of VLDL, LDL, and HDL subclasses, Lp(a),
and patterns of LDL size (LDL subclass A, AB, or B) (46).
GGE. The GGE also determines the size and estimates
the relative concentration of lipoprotein subclasses. Seven
different LDL subclasses are identified by GGE and their
concentration is estimated by densitometry. Such testing is
available from the Berkeley HeartLab (Burlingame,
California).
In the Stanford Study (47) and the Physicians’ Health
Study (48), small, dense LDL by GGE predicted CAD but
was not independent of the TC/HDL-C ratio and nonfast-
ing TG, respectively. The SCRIPS (Stanford Coronary
Risk Intervention Project) (49) trial, a 4-year angiographic
trial, found that small, dense LDL by GGE predicted both
the progression of stenosis in the control group and the vast
majority of angiographic benefit in the group on multifac-
torial intervention. Those with predominantly large LDL,
despite the same LDL-C reduction as those with small,
dense LDL, showed angiographic progression at the same
rate as the usual care (control) group.
Treatment Implications of LDL Heterogeneity
Although hyper-TG and reduced HDL-C, commonly ob-
served in diabetes (16) and the metabolic syndrome (17), are
Mudd et al. 1739
LDL Heterogeneity in CAD
reflected in the conventional lipid profile, calculated
LDL-C often does not reflect small, dense LDL-P in such
patients (13). Recognition of this is important, because
increased small, dense LDL-P due to enhanced production
of TG-rich VLDL appears to be the single most frequent
dyslipidemia in patients with premature CAD (13–16).
Thus, a low or normal LDL-C level may result in a less
aggressive approach than may be clinically indicated, par-
ticularly in the patient who may benefit more from com-
bined lipid-altering therapy.
Multiple biochemical and cellular mechanisms of athero-
genic small, dense LDL particles have been proposed (16) (Fig.
2). It is still incompletely resolved whether risk of CAD is
simply related to LDL-P per se, or whether LDL size is also
an independent predictor of CAD (see preceding text). Al-
though studies have demonstrated improvement in LDL
particle size distribution with pharmacologic intervention, it
remains to be established whether risk stratification based
on LDL size by “advanced” or additional lipoprotein testing
is sufficiently additive to merit a change in management
strategies.
Altering LDL Particle Number and Size
Statins are the most potent class of drugs to reduce LDL-P,
including small, dense LDL-P. Higher doses of the more
powerful statins may also decrease VLDL production,
leading to a small shift from small, dense to larger LDL
particles. Both niacin and fibrates convert small, dense LDL
into larger, more buoyant LDL.
Although LDL-P clearly predicts CAD events, it is
unclear if altering LDL size decreases CAD events inde-
pendent of any effect on lowering LDL-P. In the VA-HIT
(Veterans Affairs High-Density Lipoprotein Intervention
Trial) study (50) subjects with low HDL-C and average
LDL-C treated with gemfibrozil had a significant decrease
in CAD events compared with placebo, an effect propor-
tional to the increase in HDL-C, because LDL-C was not
decreased with gemfibrozil. Using NMR, Otvos et al. (51)
reported that gemfibrozil lowered small, dense LDL-P,
thereby increasing total LDL size. Total HDL-P, particu-
larly HDL-3, increased with gemfibrozil. These changes in
particle composition independently decreased risk of new
CAD events and explained a significant amount of the
benefit of gemfibrozil in the VA-HIT study.
Additionally, the DAIS (Diabetes Atherosclerosis In-
tervention Study) trial (52) showed that patients with
type 2 diabetes treated with fenofibrate had a significantly
decreased rate of angiographic CAD progression and a
greater increase in LDL size compared with placebo
controls. In a substudy, Vakkilainen et al. (53) found that
small LDL size was associated with greater CAD pro-
gression, regardless of treatment group, and that fenofi-
brate produced a greater increase in LDL particle size
compared with placebo. In contrast, the FIELD (Feno-
1740 Mudd et al.
LDL Heterogeneity in CAD
fibrate Intervention and Event Lowering in Diabetes) study
(54) was associated with a nonsignificant 11% reduction in the
primary end point (coronary heart disease death and non-
fatal myocardial infarction); however, changes in LDL-P
and LDL particle size were not reported.
Perspectives and Future Directions
The evidence supporting the incremental value of measur-
ing apoB is strong and internally consistent. The Thirty-
Person/Ten-Country Panel has proposed that evidenced-
based guidelines endorse alternatives such as apoB in both
men and women (28), a recommendation further supported
by the results of the large Copenhagen City Heart Study
(31). Statin-treated patients may have a proportionally
higher apoB than LDL-C, indicating that too many
LDL-P are still present. Those patients will require further
intervention, often in the form of combined lipid-altering
drugs. For patients with CAD, CAD risk equivalents, or a
high risk of CAD, a more aggressive LDL-C goal of ⬍70
mg/dl should be considered. This goal is in fact less than the
fifth percentile for LDL-C. The recently proposed goal
of ⬍80 mg/dl for apoB in such high-risk patients is about
the 25th percentile. Although it is not certain what the
equivalent goal of apoB should be when the LDL-C goal is
⬍70 mg/dl, the equivalent percentile for apoB is ⬍60 mg/dl.
Recent evidence indicates that non–HDL-C and LDL-P
also provide additional prognostic information over
LDL-C. Population-based percentiles for LDL-C, non–
HDL-C, apoB, and LDL-P are available from the Fra-
mingham Heart Study (36). Alternative goals for LDL-P as
well as apoB related to current NCEP treatment goals for
LDL-C and non–HDL-C should be considered in future
recommendations.
Further study is needed to define the incremental value of
LDL-P and LDL size in the setting of LDL-C targets ⬍70
mg/dl. The preponderance of small, dense LDL-P may
provide independent information that predicts risk beyond
the traditional lipid profile, particularly in the presence of
the dyslipidemic triad. With regard to high LDL-P and low
HDL, there is evidence that the ratio of apoB/apoA-I may
be the strongest predictor of CAD and that such measure-
ment could be incorporated into clinical practice.
Population-based percentiles for men and women for
apoA-I and HDL-C also are available from the Framing-
ham Heart Study (36).
Only through well-designed prospective clinical trials will
we be able to determine when lipoprotein heterogeneity
analysis is justified and how best to use the information. For
now, selective measurement of apoB might be incorporated
into clinical practice (when the TG level is above normal),
with the addition of the apoB/apoA-I ratio and LDL-P
only when it is likely to change clinical management.
JACC Vol. 50, No. 18, 2007
October 30, 2007:1735–41
Reprint requests and correspondence: Dr. Peter O. Kwiterovich,
Jr., University Lipid Clinic, Suite 310, 550 North Broadway
Building, Baltimore, Maryland 21205. E-mail: pkwitero@
jhmi.edu.
REFERENCES
1. Adult Treatment Panel III. Executive summary of the third report of
the National Cholesterol Education Program (NCEP) Expert Panel
on Detection, Evaluation, and Treatment of High Blood Cholesterol
in Adults. JAMA 2001;285:2486 –97.
2. Grundy SM, Cleeman JI, Merz CN, et al. Implications of recent
clinical trials for the National Cholesterol Education Program Adult
Treatment Panel III guidelines. Circulation 2004;110:227–39.
3. O’Keefe JH Jr., Cordain L, Harris WH, Moe RH, Vogel R. Optimal
low-density lipoprotein is 50 to 70 mg/dl. Lower is better and
physiologically normal. J Am Coll Cardiol 2004;43:2142– 46.
4. Robinson JG, Smith B, Maheshwari N, Schrott H. Pleiotrophic effects
of statins: benefit beyond cholesterol reduction? A meta-regression
analysis. J Am Coll Cardiol 2005;46:1855– 62.
5. Heart Protection Study Collaborative Group. MRC/BHF heart pro-
tection study of cholesterol lowering with simvastatin in 20,536
high-risk individuals: a randomized placebo-controlled trial. Lancet
2002;360:7–22.
6. PROVE-IT Investigators. Intensive versus moderate lipid lowering with
statins after acute coronary syndromes. N Engl J Med 2004;350:1495–
504.
7. REVERSAL Investigators. Effect of intensive compared with mod-
erate lipid-lowering therapy on progression of coronary atherosclerosis:
a randomized control study. JAMA 2004;291:1071– 80.
8. LaRosa JC, Grundy SM, Waters DD, et al. Intensive lipid lowering
with atorvastatin in patients with stable coronary artery disease.
N Engl J Med 2005;352:1425–35.
9. Nissen SE, Nicholls SJ, Sipahi I, et al., ASTEROID Investigators.
Effect of very-high intensity statin therapy on regression of coro-
nary atherosclerosis. The ASTEROID trial. JAMA 2006;295:
1556 – 65.
10. Pedersen TR, Faergeman O, Kastelein JJ, et al. High-dose atorvastatin
vs usual-dose simvastatin for secondary prevention after myocardial
infarction. The IDEAL study: a randomized controlled trial. JAMA
2005;294:2437– 45.
11. Law MR, Wald NJ, Rudnicka AR. Quantifying effect of statins on low
density lipoprotein cholesterol, ischaemic heart disease, and stroke;
systematic review and meta-analysis. BMJ 2003;326:1–7.
12. Otvos JD, Jeyarajah EJ, Cromwell WC. Measurement issues related to
lipoprotein heterogeneity. Am J Cardiol 2002;90 Suppl:22i–9i.
13. Sniderman AD. How, when, and why to use apolipoprotein B in
clinical practice. Am J Cardiol 2002;90 Suppl:48i–54i.
14. Kwiterovich PO. Clinical relevance of the biochemical, metabolic and
genetic factors that influence low-density lipoprotein heterogeneity.
Am J Cardiol 2002;90 Suppl:30i– 45i.
15. Blake GJ, Otvos JD, Rifai N, Ridker PJ. Low-density lipoprotein
particle concentration and size as determined by NMR spectroscopy as
predictors of cardiovascular disease in women. Circulation 2002;106:
1930 –7.
16. Sniderman AD, Scantlebury T, Cianflone K. Hypertriglyceridemic
hyperapoB: the unappreciated atherogenic dyslipoproteinemia in type
2 diabetes. Ann Intern Med 2001;135:447–59.
17. Kathiresan S, Otvos JD, Sullivan ML, et al. Increased small low-
density lipoprotein particle number: a prominent feature of the
metabolic syndrome in the Framingham Heart Study. Circulation
2006;113:20 –9.
18. Friedewald WT, Levy RI, Fredrickson DS. Estimation of low-density
lipoprotein in plasma without use of preparative ultracentrifuge. Clin
Chem 1972;18:499 –502.
19. Scharnagl H, Nauck M, Wieland H, Marz W. The Friedewald
formula underestimates LDL-C cholesterol at low concentrations.
Clin Chem Lab Med 2001;39:426 –31.
20. Frost PH, Havel RJ. Rationale for use of nonhigh-density lipoprotein
cholesterol rather than low-density lipoprotein cholesterol as a tool for
JACC Vol. 50, No. 18, 2007
October 30, 2007:1735–41
lipoprotein cholesterol screening and assessment of risk and therapy.
Circulation 1998;97:1837– 47.
21. Cui Y, Blumenthal RS, Flaws JA, et al. Non-high-density lipoprotein
cholesterol level as a predictor of cardiovascular disease mortality. Arch
Intern Med 2001;161:1413–9.
22. Harchaoui KE, van der Steeg WA, Stroes ESG, et al. Value of
low-density lipoprotein particle number and size as predictors of
coronary artery disease in apparently healthy men and women. The
EPIC-Norfolk prospective population study. J Am Coll Cardiol
2007;49:547–53.
23. Liu J, Sempos CT, Donahue RP, Dorn J, Trevisan M, Grundy SM.
Non-high-density lipoprotein and very-low-density lipoprotein cho-
lesterol and their risk predictive values in coronary heart disease. Am J
Cardiol 2006;98:1363– 8.
24. Marcovina SM, Albers JJ, Kennedy H, Mei JV, Henderson LO,
Hannon WH. International Federation of Clinical Chemistry stan-
dardization project for measurements of apolipoproteins A-I and B.
IV. Comparability of apolipoprotein B values by use of international
reference material. Clin Chem 1994;40:586 –92.
25. Lamarche B, Tchernof A, Moorjani S, et al. Small, dense low-density
lipoprotein particles as a predictor of the risk of ischemic heart disease
in men. Prospective results from the Quebec Cardiovascular Study.
Circulation 1997;95:69 –75.
26. St-Pierre A, Cantin B, Dagenais GR, et al. Low-density lipoprotein
subfractions and the long-term risk of ischemic heart disease in men.
13-year follow-up data from the Quebec Cardiovascular Study. Arte-
rioscler Thromb Vasc Biol 2005;25:553–9.
27. Pischon T, Girman CJ, Sacks FM, et al. Non-high-density lipoprotein
cholesterol and apolipoprotein B in the prediction of coronary heart
disease in men. Circulation 2005;112:3375– 83.
28. Barter PJ, Ballantyne CM, Carmena R, et al. Apo B versus
cholesterol in estimating cardiovascular risk and in guiding therapy:
report of the Thirty-Person/Ten-Country Panel. J Intern Med
2006;259:247–58.
29. Walldius G, Jungner I, Holme I, et al. High apolipoprotein B, low
apolipoprotein A-I, and improvement in the prediction of fatal
myocardial infarction (AMORIS study): a prospective study. Lancet
2001;358:2026 –33.
30. Moss AJ, Goldstein RE, Marder VJ, et al. Thrombogenic factors and
recurrent coronary events. Circulation 1999;99:2517–22.
31. Benn M, Nordestgaard BG, Jensen GB, Tybjaerg-Hansen A. Improv-
ing prediction of ischemic cardiovascular disease in the general
population using apolipoprotein B. The Copenhagen City Heart
Study. Arterioscler Thromb Vasc Biol 2007;27:661–70.
32. Gotto AM Jr., Whitney E, Stein EA, et al. Relation between baseline
and on-treatment lipid parameters and first acute major coronary
events in the Air Force/Texas Coronary Atherosclerosis Prevention
Study (AFCAPS/TEXCAPS). Circulation 2000;101:477– 84.
33. van Lennep JE, Westerveld HT, van Roeters Lennep HWO, et al.
Apolipoprotein concentrations during treatment and recurrent coro-
nary artery disease events. Arterioscler Thromb Vasc Biol 2000;20:
2408 –13.
34. Schaefer EJ, Lamon-Fava S, Cohn SD, et al. Effects of age, gender,
and menopausal status on plasma low density lipoprotein cholesterol
and apolipoprotein B levels in the Framingham Offspring Study. J
Lipid Res 1994;35:779 –92.
35. Schaefer EJ, Lamon-Fava S, Ordovas JM, et al. Factors associated
with low and elevated plasma high density lipoprotein cholesterol and
apolipoprotein A-I levels in the Framingham Offspring Study. J Lipid
Res 1994;35:871– 82.
36. Freedman DS, Otvos JD, Jeyarajah EJ, et al. Sex and age differences in
lipoprotein subclasses measured by nuclear magnetic resonance spec-
troscopy: the Framingham Study. Clin Chem 2004;50:1189 –200.
37. Kuller L, Arnold A, Tracy R, et al. Nuclear magnetic resonance
spectroscopy of lipoproteins and risk of coronary heart disease in the
Mudd et al. 1741
LDL Heterogeneity in CAD
Cardiovascular Health Study. Arterioscler Thromb Vasc Biol 2002;
22:1175– 80.
38. Benton JL, Blumenthal RS, Becker DM, Yanek LR, Moy TF, Post
W. Predictors of low-density lipoprotein particle size in a high-risk
African American population. Am J Cardiol 2005;95:1320 –32.
39. Ensign W, Hill N, Heward CB. Disparate LDL phenotypic classifi-
cation among 4 different methods assessing LDL particle characteris-
tics. Clin Chem 2006;52:1722–7.
40. Mackey RH, Kuller LH, Sutton-Tyrell K, et al. Lipoprotein subclasses
and coronary artery calcium in postmenopausal women from the
Healthy Women Study. Am J Cardiol 2002;90:71i– 6i.
41. Rosenson RS, Otvos JD, Freedman DS. Relations of lipoprotein
subclass levels and low-density lipoprotein size to progression of
coronary artery disease in the Pravastatin Limitation of Atherosclerosis
in the Coronary Arteries (PLAC-1) trial. Am J Cardiol 2002;90:
89 –94.
42. Mora SA, Szklo M, Otvos JD, et al. LDL particle subclasses, LDL
particle size, and carotid atherosclerosis in the Multi-Ethnic Study of
Atherosclerosis (MESA). Atherosclerosis 2006;192:211–7.
43. Watts GF, Mandalia S, Brunt JN, Slavin BM, Coltart DJ, Lewis B.
Independent associations between plasma lipoprotein subfraction lev-
els and the course of coronary artery disease in the St. Thomas’
Atherosclerosis Regression Study (STARS). Metabolism 1993;42:
1461–7.
44. Mack WJ, Krauss RM, Hodis HN. Lipoprotein subclasses in the
Monitored Atherosclerosis Regression Study (MARS). Treatment
effects and relation to coronary angiographic progression. Arterioscler
Thromb Vasc Biol 1996;16:697–704.
45. Zambon A, Hokanson JE, Brown BG, Brunzell JD. Evidence for a
new pathophysiological mechanism for coronary artery disease regres-
sion: hepatic lipase-mediated changes in LDL-C density. Circulation
1999;99:1959 – 64.
46. Kulkami KR, Garber DW, Marcovina SM, Segrest JP. Quantification
of cholesterol in all lipoprotein classes by the VAP-II method. J Lipid
Res 1994;35:159 – 68.
47. Gardner CD, Fortmann SP, Krauss RM. Association of small low-
density lipoprotein particles with the incidence of coronary artery
disease in men and women. JAMA 1996;276:875– 81.
48. Stampfer MJ, Krauss RM, Ma J, et al. A prospective study of TG level,
low-density lipoprotein particle diameter, and risk of myocardial
infarction. JAMA 1996;76:882– 8.
49. Williams PT, Superko HR, Haskell WL, et al. Smallest LDL particles
are most strongly related to coronary disease progression in men.
Arterioscler Thromb Vasc Biol 2003;23:314 –21.
50. Robins SJ, Collins D, Wittes JT, et al. Relation of gemfibrozil
treatment and lipid levels with major coronary events: VA-HIT: a
randomized control trial. JAMA 2001;285:1585–91.
51. Otvos JD, Collins D, Freedman DS, et al. Low-density lipoprotein
and high-density lipoprotein particle subclasses predict coronary events
and are favorably changed by gemfibrozil therapy in the Veterans
Affairs High-Density Lipoprotein Intervention. Circulation 2006;113:
1556 – 63.
52. Diabetes Atherosclerosis Intervention Study Investigators. Effect of
fenofibrate on progression of coronary-artery disease in type 2 diabetes:
the Diabetes Atherosclerosis Intervention Study, a randomized study.
Lancet 2001;357:905–10.
53. Vakkilainen J, Steiner G, Ansquer J-C, et al. Relationships between
low-density lipoprotein particle size, plasma lipoproteins, and progres-
sion of coronary artery disease. The Diabetes Atherosclerosis Inter-
vention Study (DAIS). Circulation 2003;107:1733–7.
54. Keech A, Simes RJ, Barter P, et al., FIELD Study Investigators.
Effects of long-term fenofibrate therapy on cardiovascular events in
9795 people with type 2 diabetes (the FIELD study): randomized
placebo-controlled trial. Lancet 2003;361:2005–16.