NCEP Report

Implications of Recent Clinical Trials for the

National Cholesterol Education Program
Adult Treatment Panel III Guidelines
Scott M. Grundy; James I. Cleeman; C. Noel Bairey Merz; H. Bryan Brewer, Jr; Luther T. Clark;
Donald B. Hunninghake*; Richard C. Pasternak; Sidney C. Smith, Jr; Neil J. Stone;
for the Coordinating Committee of the National Cholesterol Education Program
Endorsed by the National Heart, Lung, and Blood Institute, American College of Cardiology Foundation,
and American Heart Association
Abstract—The Adult Treatment Panel III (ATP III) of the National Cholesterol Education Program issued an
evidence-based set of guidelines on cholesterol management in 2001. Since the publication of ATP III, 5 major clinical
trials of statin therapy with clinical end points have been published. These trials addressed issues that were not examined
in previous clinical trials of cholesterol-lowering therapy. The present document reviews the results of these recent trials
and assesses their implications for cholesterol management. Therapeutic lifestyle changes (TLC) remain an essential
modality in clinical management. The trials confirm the benefit of cholesterol-lowering therapy in high-risk patients and
support the ATP III treatment goal of low-density lipoprotein cholesterol (LDL-C) ⬍100 mg/dL. They support the
inclusion of patients with diabetes in the high-risk category and confirm the benefits of LDL-lowering therapy in these
patients. They further confirm that older persons benefit from therapeutic lowering of LDL-C. The major recommen-
dations for modifications to footnote the ATP III treatment algorithm are the following. In high-risk persons, the
recommended LDL-C goal is ⬍100 mg/dL, but when risk is very high, an LDL-C goal of ⬍70 mg/dL is a therapeutic
option, ie, a reasonable clinical strategy, on the basis of available clinical trial evidence. This therapeutic option extends
also to patients at very high risk who have a baseline LDL-C ⬍100 mg/dL. Moreover, when a high-risk patient has high
triglycerides or low high-density lipoprotein cholesterol (HDL-C), consideration can be given to combining a fibrate or
nicotinic acid with an LDL-lowering drug. For moderately high-risk persons (2⫹ risk factors and 10-year risk 10% to
20%), the recommended LDL-C goal is ⬍130 mg/dL, but an LDL-C goal ⬍100 mg/dL is a therapeutic option on the
basis of recent trial evidence. The latter option extends also to moderately high-risk persons with a baseline LDL-C of
100 to 129 mg/dL. When LDL-lowering drug therapy is employed in high-risk or moderately high-risk persons, it is
advised that intensity of therapy be sufficient to achieve at least a 30% to 40% reduction in LDL-C levels. Moreover,
any person at high risk or moderately high risk who has lifestyle-related risk factors (eg, obesity, physical inactivity,
elevated triglycerides, low HDL-C, or metabolic syndrome) is a candidate for TLC to modify these risk factors
regardless of LDL-C level. Finally, for people in lower-risk categories, recent clinical trials do not modify the goals and
cutpoints of therapy. (J Am Coll Cardiol 2004;44:720-32)
Key Words: cholesterol 䡲 trials 䡲 lipoproteins 䡲 coronary disease

T he Executive Summary of the Third Report of the National

Cholesterol Education Program (NCEP) Expert Panel on
Detection, Evaluation, and Treatment of High Blood Cholesterol
mendations on the management of high blood cholesterol and
related disorders. For development of its recommendations, ATP
III places primary emphasis on large, randomized, controlled
in Adults (Adult Treatment Panel III [ATP III]) was published in clinical trials (RCTs). In the past decade, a series of large RCTs
May 2001 (1). The full report of ATP III was published in have yielded a vast body of data for these recommendations.
December 2002 (2). ATP III provides evidence-based recom- Other lines of evidence, including prospective epidemiological

This statement was approved by the National Heart, Lung, and Blood Institute in April 2004, by the American Heart Association Science Advisory
and Coordinating Committee in May 2004, and by the American College of Cardiology Foundation Board of Trustees in May 2004. A single reprint is
available by calling 800-242-8721 (US only) or writing the American Heart Association, Public Information, 7272 Greenville Ave, Dallas, TX
75231-4596. Ask for reprint No. 71-0292. To purchase additional reprints: up to 999 copies, call 800-611-6083 (US only) or fax 413-665-2671; 1000
or more copies, call 410-528-4121, fax 410-528-4264, or e-mail kgray@lww.com. To make photocopies for personal or educational use, call the
Copyright Clearance Center, 978-750-8400.
*Member of the Working Group until December 31, 2003.
© 2004 by the American College of Cardiology Foundation and the American Heart Association, Inc.

Disclosure information for the members of the working group that drafted the ATP III update is available on the NHLBI website
(www.nhbli.nih.gov/guidelines/cholesterol/atp3upd04.htm).

doi:10.1016/j.jacc.2004.07.001
JACC Vol. 44, No. 3, 2004 Grundy et al. 721
August 4, 2004:720–32 Recent Clinical Trials and NCEP ATP III

studies and smaller clinical trials, afford additional evidence for an LDL-lowering drug was said to be optional. Alternatively,
crafting the recommendations. if the patient has elevated triglycerides or low high-density
All ATP reports have identified low-density lipoprotein lipoprotein cholesterol (HDL-C), a drug that targets these
cholesterol (LDL-C) as the primary target of cholesterol- abnormalities may be added.
lowering therapy. Many prospective studies have shown that Compared with ATP II (3), ATP III added new intensity to
high serum concentrations of LDL-C are a major risk factor LDL-C lowering in patients with multiple (2⫹) CHD risk
for coronary heart disease (CHD). A large number of RCTs, factors. Previous ATP guidelines established the LDL-C goal
moreover, have documented that lowering of LDL-C levels for this category to be a level ⬍130 mg/dL. This goal was
will reduce the risk for major coronary events. In ATP II (3), retained in ATP III, but risk assessment was expanded
evidence for the benefit of LDL-lowering therapy was based beyond the counting of risk factors. ATP III recommended
on analysis and meta-analysis of RCTs that were carried out that Framingham risk scoring be carried out in individuals
with therapies other than HMG CoA reductase inhibitors with 2⫹ risk factors so as to triage them into 3 levels of
(statins). ATP III (1,2) reviewed new data from 5 large RCTs 10-year risk for hard CHD events (myocardial infarction ⫹
with statins. Results of several smaller RCTs with statins and CHD death): ⬎20%, 10% to 20%, and ⬍10%. Persons with
other drugs also were examined. On the basis of accumulated a 10-year risk ⬎20% were elevated to the high-risk category;
evidence from epidemiological studies and RCTs, ATP III for them, the LDL-C goal is ⬍100 mg/dL. For others with 2⫹
proposed a treatment algorithm for LDL-lowering therapy. risk factors and a 10-year risk ⱕ20%, the LDL-C goal is
Since the publication of ATP III, 5 major clinical trials ⬍130 mg/dL. LDL-lowering dietary therapy is universally
with statin therapy and clinical end points have been pub- advocated for patients with an LDL-C above the goal level. If
lished. These include the Heart Protection Study (HPS) (4), the 10-year risk is 10% to 20%, drug therapy should be
the Prospective Study of Pravastatin in the Elderly at Risk considered if the LDL-C level is above the goal level (ie,
(PROSPER) (5), Antihypertensive and Lipid-Lowering ⱖ130 mg/dL) after a trial of dietary therapy. When 10-year
Treatment to Prevent Heart Attack Trial—Lipid-Lowering risk is ⬍10%, an LDL-lowering drug can be considered if the
Trial (ALLHAT-LLT) (6), Anglo-Scandinavian Cardiac Out- LDL-C level is ⱖ160 mg/dL on maximal dietary therapy.
comes Trial—Lipid-Lowering Arm (ASCOT-LLA) (7), and Finally, most persons with 0 to 1 risk factor have a 10-year
the Pravastatin or Atorvastatin Evaluation and Infection— risk ⬍10%. For these individuals, clinical management and
Thrombolysis in Myocardial Infarction 22 (PROVE IT–TIMI dietary therapy is recommended when the LDL-C level is
22) trial (8). These trials addressed issues that had not been ⱖ160 mg/dL. The goal is to lower LDL-C concentrations to
adequately addressed in previous statin trials. The results ⬍160 mg/dL. If the LDL-C is ⱖ190 mg/dL after an adequate
appear to have important implications for the management of trial of dietary therapy, consideration should be given to
patients with lipid disorders, particularly for high-risk pa- adding a cholesterol-lowering drug. When serum LDL-C
tients. They further may require some rethinking of the ranges from 160 to 189 mg/dL, introduction of a cholester-
treatment thresholds of ATP III recommendations. In addi- ol-lowering drug is a therapeutic option in appropriate cir-
tion, findings of other smaller trials or subgroup analyses of cumstances, such as when a severe risk factor is present. ATP
major trials have been published. The purpose of the present III outlines several factors that can be taken into consider-
document is to examine the results of all of these studies and ation to guide clinical judgment for this category.
to assess their implications in relation to the ATP III report. ATP III placed major emphasis on therapeutic lifestyle
First, we will summarize the principal elements of the ATP changes (TLC) as an essential modality in clinical manage-
III treatment algorithm and the major findings of the recent
ment for persons at risk for cardiovascular disease (CVD).
trials.
ATP III’s TLC approach was designed to achieve risk
According to the ATP III algorithm, persons are catego-
reduction through both LDL-C lowering and metabolic syn-
rized into 3 risk categories: (1) established CHD and CHD
drome management. Therefore, when the implications of
risk equivalents, (2) multiple (2⫹) risk factors, and (3) zero to
recent LDL-lowering drug trials are considered, it must be
one (0 –1) risk factor. CHD risk equivalents include noncor-
reemphasized that the results do not in any way diminish the
onary forms of clinical atherosclerotic disease, diabetes, and
importance of lifestyle change for CVD risk reduction.
multiple (2⫹) CHD risk factors with 10-year risk for CHD
⬎20%. All persons with CHD or CHD risk equivalents can
Review of Recent Clinical Trials With Major
be called high risk. The goal for LDL-lowering therapy in
high-risk patients is an LDL-C level ⬍100 mg/dL. According
Cardiovascular End Points
to ATP III, for a baseline or on-treatment LDL-C ⬍100 Heart Protection Study
mg/dL, no further LDL-lowering therapy was recommended. This clinical trial was carried out in 20 536 adults living in the
For all high-risk patients with LDL-C levels ⱖ100 mg/dL, United Kingdom (aged 40 to 80 years) who were at high risk
LDL-lowering dietary therapy should be initiated. When for a CVD event (4). Entrance criteria included coronary
baseline LDL-C is ⱖ130 mg/dL, an LDL-lowering drug disease, other occlusive arterial disease, or diabetes. Patients
should be started simultaneously with dietary therapy. How- were randomly allocated to 40 mg simvastatin daily or
ever, LDL-lowering drugs were not mandated if the baseline placebo. Primary outcomes included total mortality for over-
LDL-C level is in the range of 100 to 129 mg/dL; in this all analysis and fatal or nonfatal vascular events for subcat-
range, ATP III suggested several therapeutic options. Dietary egory analyses. The incidence of cancer and other major
therapy should be intensified, whereas adding or intensifying morbidity also was determined.
722 Grundy et al. JACC Vol. 44, No. 3, 2004
Recent Clinical Trials and NCEP ATP III August 4, 2004:720–32

Serum lipids at baseline were determined on nonfasting HPS investigators further examined their results more
samples. Levels of LDL-C were measured by the direct LDL closely for persons with diabetes (11). The study included
method (9). Average lipid values at baseline were total 5963 individuals with diabetes (ages 40 to 80 years). Those
cholesterol 228 mg/dL, triglycerides 186 mg/dL (nonfasting), subjects receiving simvastatin 40 mg/d had significant reduc-
HDL-C 41 mg/dL, non-HDL-C 187 mg/dL, and direct tions of approximately one quarter in first-event rates for
LDL-C 131 mg/dL. In most other clinical trials of cholester- major coronary events, strokes, and revascularizations. Event
ol-lowering therapy, serum lipid levels have been determined reductions were similar to those for nondiabetic patients. In
on fasting samples, and LDL-C has been calculated by the 2912 patients with diabetes and without diagnosed coronary
Friedewald equation [LDL-C ⫽ total cholesterol ⫺ HDL-C or other occlusive arterial disease at entry, simvastatin ther-
⫺ VLDL-C (triglycerides/5)], where VLDL indicates very- apy reduced risk by about one third. In 2426 participants with
low-density lipoprotein (10). This calculation includes diabetes whose pretreatment LDL-C was ⬍116 mg/dL, event
intermediate-density lipoprotein in the LDL fraction. If this rates were 27% lower on simvastatin therapy. In the subgroup
equation were applied to the HPS values cited above, the of patients with diabetes who were without vascular disease
average calculated LDL-C would be approximately 150 and whose LDL-C levels were ⬍116 mg/dL at baseline, a
mg/dL [(228⫺41⫺(186/5)]. However, because the baseline marginally significant 30% reduction in risk was observed.
samples were nonfasting, the triglyceride levels were likely to Efficacy of simvastatin therapy in the subgroup of patients
have been at least 20 to 30 mg/dL higher than fasting; with LDL-C ⬍100 mg/dL was not reported. HPS investiga-
consequently, applying the Friedewald equation to baseline tors concluded that, in general, cholesterol lowering with
levels would underestimate LDL-C by 4 to 6 mg/dL [(20 to statin therapy is efficacious in patients with diabetes, includ-
30 mg/dL)/5] because this much cholesterol was falsely ing those without manifest CHD and those with relatively low
attributed to VLDL-C. Consequently, estimations of baseline LDL-C levels.
fasting LDL-C, if calculated by the Friedewald equation,
likely would have been in the range of 150 to 155 mg/dL, or Prospective Study of Pravastatin in the Elderly
about 15% higher than baseline LDL-C calculated by the at Risk
direct method. If this difference between direct and calculated This trial examined the efficacy of pravastatin treatment in
LDL-C holds at low LDL-C, a direct LDL-C level of 100 older men and women with or at high risk of developing CVD
mg/dL would correspond to a calculated LDL-C of 115 and stroke (5). Subjects (n⫽5804; 2804 men and 3000
mg/dL. Although this difference could be of some signifi- women), ages 70 to 82 years, who had a history of vascular
cance for treatment decisions, to avoid confusion the distinc- disease or CVD risk factors were randomized to pravastatin
tion will not be emphasized in the discussion to follow. (40 mg/d) or placebo. The primary end point was a composite
In patients allocated to simvastatin, all-cause mortality was of coronary death, nonfatal myocardial infarction, and fatal or
significantly reduced by 13% (P⫽0.0003). Major vascular nonfatal stroke. Baseline total cholesterol varied widely from
events were reduced by 24%, coronary death rate by 18%, 150 mg/dL to 350 mg/dL. Follow-up averaged 3.2 years.
nonfatal myocardial infarction ⫹ coronary death by 27%, Pravastatin reduced LDL-C levels by 34%. The composite
nonfatal or fatal stroke by 25%, and cardiovascular revascu- end point was reduced on pravastatin therapy by 15%
larization by 24%. The reduction in the event rate was similar (P⫽0.014). Major coronary events, defined as nonfatal myo-
in each subcategory, including patients without diagnosed cardial infarction and CHD death, fell on therapy by 19%
coronary disease who had cerebrovascular disease, or periph- (P⫽0.006), and CHD mortality by 24% (P⫽0.043). No
eral artery disease, or diabetes. Similar event reductions on reduction in stroke was observed, but transient ischemic
simvastatin therapy occurred for men and women and for attacks fell by 25% on therapy (P⫽0.051). The stroke rate in
participants either under or over 70 years of age at entry. No the trial, however, was about half of that predicted, so the
significant adverse effects of simvastatin therapy were re- effects of statin therapy on stroke must be viewed in this light.
ported, including no significant increase in myopathy, cancer New cancer unexpectedly was found 25% more often on
incidence, or hospitalization for any other nonvascular cause. pravastatin treatment (P⫽0.020). This finding, however,
Subgroup analysis of HPS suggests that simvastatin ther- contrasts with meta-analysis of all pravastatin and all statin
apy produced similar reductions in relative risk regardless of trials, in which overall cancer incidence was not increased
the baseline levels of LDL-C, including subgroups with initial (5). Pravastatin therapy neither improved cognitive function
(or baseline) LDL-C levels ⱖ135 mg/dL, ⬍116 mg/dL, or nor retarded progression of disability. According to the
⬍100 mg/dL. At least 2 issues, however, can be noted with authors, PROSPER results allow statin therapy to be extended
regard to the reported subgroup analysis of HPS at low (or to older persons.
very low) LDL-C levels. First, LDL-C cutpoints to define
these subgroups would have been higher if LDL-C had been Antihypertensive and Lipid-Lowering
calculated by the Friedewald equation, the method employed Treatment to Prevent Heart Attack
by ATP III for routine clinical practice. Second, the charac- Trial—Lipid-Lowering Trial
teristics of low-LDL subgroups, ie, what portions had hyper- The primary goal of ALLHAT was to evaluate current
triglyceridemia, elevated non-HDL-C, or diabetes, or were modalities of hypertension treatment. The lipid-lowering
free of CVD, have not been made available. These qualifying component, which was a subset of this trial, was designed to
issues must be kept in mind when generalizing HPS findings assess whether pravastatin therapy compared with usual care
to all high-risk patients with low baseline LDL-C levels. reduces all-cause mortality in older, moderately hypercholes-
JACC Vol. 44, No. 3, 2004 Grundy et al. 723
August 4, 2004:720–32 Recent Clinical Trials and NCEP ATP III

terolemic, hypertensive participants with at least one addi- events had occurred in the atorvastatin group, compared with
tional CHD risk factor (6). The study used 513 primarily 154 events in the placebo group (hazard ratio 0.64,
community-based North American clinical centers. The lipid- P⫽0.0005). In the atorvastatin group, incidence of fatal and
lowering component of ALLHAT randomized 10 355 per- nonfatal stroke was reduced by 27% (P⫽0.024), total cardio-
sons. Participants were over 55 years of age and had LDL-C vascular events by 21% (P⫽0.0005), and total coronary
levels ranging from 120 to 189 mg/dL and triglycerides events by 29% (P⫽0.0005). There was a nonsignificant trend
below 350 mg/dL. Those patients with LDL-C levels ⱖ120 toward a reduction in total mortality in the atorvastatin group
mg/dL (100 to 129 mg/dL if known CHD) and triglycerides (13%; P⫽0.16). Because of these markedly positive findings
lower than 350 mg/dL were randomized to nonblinded arms with atorvastatin therapy, the study was terminated prema-
of pravastatin (n⫽5170) or usual care (n⫽5185). Baseline turely. The authors indicated that LDL lowering with atorva-
mean total cholesterol was 224 mg/dL; LDL-C, 146 mg/dL; statin therapy has considerable potential to reduce risk for
HDL-C, 48 mg/dL; and triglycerides, 152 mg/dL. Mean age CVD in primary prevention in patients with multiple CVD
was 66 years; 49% were women; 38% were black and 23% risk factors.
Hispanic; 14% had a history of CHD; and 35% had type 2
diabetes. The primary outcome was all-cause mortality, and Pravastatin or Atorvastatin Evaluation and
secondary outcomes were nonfatal myocardial infarction or
Infection—Thrombolysis in Myocardial
fatal CHD (CHD events) combined, cause-specific mortality,
Infarction 22
and cancer. This study, designated PROVE IT,(8) was designed to deter-
Mean follow-up duration of participants was 4.8 years. mine whether intensive LDL-C lowering will reduce major
Crossover of usual-care participants to lipid-lowering drugs
coronary events, including mortality, more than “standard”
was high (32% of usual-care participants with CHD and 29%
LDL-C lowering with statin therapy in high-risk patients.
without CHD). Follow-up of patients for lipid results was not
Two statins at different doses were compared: atorvastatin 80
complete. Among a nonrandom subset of participants tested,
mg versus pravastatin 40 mg. Previous studies have shown
total cholesterol levels were reduced by 17% with pravastatin
that pravastatin 40 mg produces a reduction of LDL-C
versus 8% with usual care at 4 years. In ALLHAT-LLT,
equivalent to approximately 10 mg of atorvastatin. Prior
all-cause mortality was similar for the 2 groups, with 6-year
clinical trials have demonstrated that treatment of patients
mortality rates of 14.9% for pravastatin versus 15.3% with
with established CHD with pravastatin 40 mg will reduce
usual care. For all participants, CHD event rates were not
LDL-C levels to near 100 mg/dL and will reduce risk for
significantly different between the groups, with 6-year CHD
major coronary events by approximately 27% (12). In
event rates of 9.3% for pravastatin and 10.4% for usual care.
PROVE IT, 4162 patients who had been hospitalized for an
In the African-American subgroup, however, CHD events
acute coronary syndrome within the preceding 10 days were
were significantly reduced in the pravastatin arm compared
enrolled and randomized to the 2 therapies. The primary end
with usual care. The authors speculated that the failure to
point of the trial was a composite of death from any cause,
detect a significant reduction in risk in hypertensive patients
myocardial infarction, documented unstable angina requiring
treated with pravastatin may be due to the modest differential
in total cholesterol (9.6%) between pravastatin and usual rehospitalization, revascularization (performed at least 30
care. Other possible explanations for the failure to observe a days after randomization), and stroke. Mean follow-up time
treatment benefit could be the unblinded nature of the study was 24 months. At the end of 2 years of therapy, the
without a placebo arm and a large crossover of higher-risk composite cardiovascular end point was reduced by 16% with
subjects in the usual-care arm to active lipid-lowering atorvastatin compared with pravastatin (P⬍0.005). Nonsig-
therapy. nificant trends were observed on atorvastatin therapy for total
mortality (P⬍0.07) and for death or myocardial infarction
Anglo-Scandinavian Cardiac Outcomes (P⬍0.06). The high dose of atorvastatin was well tolerated,
Trial—Lipid-Lowering Arm and no case of severe myopathy (rhabdomyolysis) was
In contrast to the ALLHAT lipid-lowering component, a observed in either treatment group. Greater than 3-fold
markedly different result was obtained in hypertensive pa- elevations of alanine aminotransferase were observed in 3.3%
tients in ASCOT-LLA (7). In this study, 19 342 hypertensive of patients treated with atorvastatin versus 1.1% on pravasta-
patients, 40 to 79 years old and having at least 3 other tin (P⬍0.003).
cardiovascular risk factors, were randomized to 1 of 2 The LDL-C level attained on pravastatin 40 mg was 95
antihypertensive regimens. Among these subjects, 10 305 mg/dL, whereas the level attained on atorvastatin 80 mg was
were in addition randomly assigned atorvastatin 10 mg or 62 mg/dL. The difference in LDL-C thus was 33 mg/dL
placebo. Selection was made on the basis of nonfasting total (35%). The results of PROVE IT suggest that more intensive
cholesterol of ⱕ251 mg/dL (6.5 mmol/L). LDL-C levels LDL-C–lowering therapy reduces major cardiovascular
averaged 132 mg/dL and were reduced by an average of 42 events in patients with acute coronary syndrome compared
mg/dL (29%) in the atorvastatin-treated group at the end of with less intensive therapy over a period of 2 years. It must be
the study. The primary end point was nonfatal myocardial noted, however, that 72% of the patients had LDL-C levels
infarction and fatal CHD. The study was planned for a ⬍125 mg/dL, and in this large subgroup, the modest trend
follow-up of an average of 5 years but was stopped after a toward benefit of atorvastatin over pravastatin was not
median follow-up of 3.3 years. At that time, 100 primary statistically significant.
724 Grundy et al. JACC Vol. 44, No. 3, 2004
Recent Clinical Trials and NCEP ATP III August 4, 2004:720–32

Lipid Targets of Therapy

LDL-C: The Primary Target of
Lipid-Lowering Therapy
The identification of an elevated LDL-C as the primary target
of lipid-lowering therapy is based on a wealth of information
from basic research, animal studies, epidemiological studies,
genetic forms of hypercholesterolemia, and controlled clini-
cal trials. Recent clinical trials add further support for the
NCEP priority on high serum LDL-C. Four new trials
(4,5,7,8) demonstrate that effective LDL-C reduction sub-
stantially reduces risk for CHD, whereas one trial (6) failed to
produce a sizable differential in LDL-C levels between
treatment and control groups and did not yield a significant Log-linear relationship between LDL-C levels and relative risk
risk reduction. for CHD. This relationship is consistent with a large body of epi-
demiological data and with data available from clinical trials of
LDL-lowering therapy. These data suggest that for every
Other Lipid Targets 30-mg/dL change in LDL-C, the relative risk for CHD is changed
It should be noted that ATP III introduced a new secondary in proportion by about 30%. The relative risk is set at 1.0 for
target of therapy, namely non-HDL-C, in patients with LDL-C⫽40 mg/dL.
elevated triglycerides (ⱖ200 mg/dL). Non-HDL-C equates to
VLDL ⫹ LDL-C (which, when calculated, includes the US Food and Drug Administration has approved one
intermediate-density lipoprotein). The non-HDL-C goal is 30 statin/nicotinic acid combination. Although the majority of
mg/dL higher than the LDL-C goal. Non-HDL-C was added patients can tolerate nicotinic acid therapy, a sizable minority
as a secondary target of therapy to take into account the are intolerant because of a variety of side effects.
atherogenic potential associated with remnant lipoproteins in
patients with hypertriglyceridemia. Because statins lower Relation of Serum LDL-C Concentrations to
LDL-C cholesterol and non-HDL-C to a similar percentage, CHD Risk
recent clinical trials do not differentiate between these 2 lipid Epidemiological surveys have shown that serum total choles-
measures with regard to their relative benefits in risk terol levels are continuously correlated with CHD risk over a
reduction. broad range of cholesterol values. This relationship has been
Although the potential benefit of HDL-raising therapy has observed in many populations throughout the world (23–25).
evoked considerable interest, current documentation of risk Because serum LDL-C levels correlate highly with total
reduction through controlled clinical trials is not sufficient to cholesterol in populations, the same relation must exist
warrant setting a specific goal value for raising HDL-C. between LDL-C concentrations and CHD risk. Although the
Recent lipid-lowering drug trials provide no new evidence in association between LDL-C levels and CHD risk is continu-
this regard (13). New drugs that effectively raise serum ous, it is not linear; risk rises more steeply with increasing
HDL-C levels are currently under development, and it is LDL-C concentrations. This results in a curvilinear, or
likely that these drugs will be tested for efficacy for clinical log-linear, relationship (Figure). This means that when the
event reduction in the future. One class of drugs that modestly relationship between LDL-C levels and CHD risk is plotted
raises HDL-C is the fibrates. Post-hoc analysis (2,14) of on a log scale, the relationship becomes linear. Thus, at any
several clinical trials with fibrates indicates that they reduce level of LDL-C, for a given milligram-per-deciliter change in
risk for CHD events in patients with high triglycerides and the LDL-C level, the change in relative risk is the same as at
low HDL-C, especially when the patients have diabetes or any other LDL-C level. This relationship has 2 important
characteristics of the metabolic syndrome (15). Although the implications. First, when persons with low LDL-C have the
evidence base to support fibrate therapy is not as strong as same absolute risk (because of other risk factors) as those
that for statins, fibrates may have an adjunctive role in the with high LDL-C, the same absolute benefit is attained for a
treatment of patients with high triglycerides/low HDL, espe- given milligram-per-deciliter lowering of LDL-C. Second,
cially in combination with statins. Concern about develop- when persons with low LDL-C have a lower absolute risk
ment of myopathy with this combination has been lessened than those with higher LDL-C, less absolute benefit is
somewhat by the recent finding that one fibrate, fenofibrate, attained for a given LDL-C lowering in the low LDL-C
does not interfere with catabolism of statins and thus likely group. Clinical trials of cholesterol-lowering therapy have
does not substantially increase the risk for clinical myopathy generally confirmed this log-linear relation. In fact, epidemi-
in patients treated with moderate doses of statins (16,17). ological studies and clinical trials have produced congruent
Another drug that raises HDL-C is nicotinic acid. Several results by showing an almost identical pattern of association
clinical trials support the efficacy of nicotinic acid for (23–25). Until recently, however, cholesterol-lowering clini-
reduction of CHD risk, both when used alone (18,19) and in cal trials in high-risk patients failed to conclusively recapit-
combination with statins (20,21). The combination of a statin ulate the relationship observed in epidemiological studies in
with nicotinic acid produces a marked reduction of LDL-C the lower ranges of LDL-C, ie, below 125 mg/dL (12). This
and a striking rise in HDL-C (22). As a result of these studies, lack of hard evidence of benefit from further reducing already
JACC Vol. 44, No. 3, 2004 Grundy et al. 725
August 4, 2004:720–32 Recent Clinical Trials and NCEP ATP III

low LDL-C concentrations made it impossible for ATP III to mg/dL does not appear to be a threshold below which no
make unequivocal recommendations on LDL-lowering ther- further benefit could be achieved by still more LDL-C
apy for persons with lower levels of serum LDL-C. lowering. It is important to note that ATP III considered an
The results of HPS help to confirm the congruence of LDL-C level of 100 mg/dL to be a minimal goal of treatment
epidemiology and clinical trials at low LDL-C levels. HPS for high-risk patients. This level was not viewed as the level
provides strong new evidence to support the log-linear of maximal benefit of LDL lowering. A goal of less than 100
relationship between LDL-C levels and CHD risk, even at mg/dL was explicitly established by ATP III to indicate that
low LDL-C concentrations. In fact, HPS results suggest that the level of 100 is a minimal goal of therapy. Both HPS and
reducing serum LDL-C from any baseline level further PROVE IT indeed suggest that additional benefit may be
lowers risk in high-risk patients. In HPS, absolute risk obtained by reducing LDL levels to substantially below 100
reductions for major vascular events were smaller at lower mg/dL. This likelihood is enhanced by the finding that
LDL-C levels because the risk imparted by higher LDL-C intensive lowering of LDL-C to well below 100 mg/dL will
itself was lacking. Nonetheless, the association between reduce progression of coronary atherosclerotic lesions com-
LDL-C levels and CHD risk seemingly remains log-linear at pared with LDL-C reductions to approximately 110 mg/dL
low LDL-C levels (Figure). The recent trials did not identify (29). If HPS is taken at face value, reducing LDL-C by 30%
a threshold LDL-C level below which no further reduction in starting at 100 mg/dL will produce another 20% to 30%
risk occurs. lowering in relative risk for CHD. In PROVE IT, the
somewhat smaller reduction of 16% in major cardiovascular
Implications of Log-Linear Relationship Between events on atorvastatin 80 mg compared with pravastatin 40
LDL-C and CHD Risk for ATP III’s Categorical mg may be related to the relatively short duration of the trial.
Goals of Therapy in High-Risk Patients Thus, in terms of absolute risk, an LDL-C of 70 mg/dL seems
Rationale for Recommended Low LDL-C preferable for high-risk patients compared with a level of 100
Goal (<100 mg/dL) mg/dL. At present, however, HPS and PROVE IT cannot be
ATP III set the goal for LDL-C lowering in high-risk patients taken as the final word on the benefit of reducing LDL levels
to be ⬍100 mg/dL. This goal is consistent with the observed to well below 100 mg/dL. Several other clinical trials (re-
log-linear relationship between LDL-C levels and CHD risk viewed in Waters et al (30)) are underway to probe the
observed in epidemiological data (23–25). It was as low as efficacy of lowering LDL to very low levels.
could be supported by clinical-trial evidence at the time of Until these trials are completed, prudence requires that
ATP III release. It also was a goal that could be achieved setting an LDL-C goal of ⬍70 mg/dL for high-risk patients
through LDL-C lowering in a sizable proportion of high-risk must be left as a therapeutic option on the basis of clinical
patients by standard doses of drugs used in clinical trials. The trial evidence, whereas a goal of ⬍100 mg/dL can be retained
latter point is important. Doses of statins used in most as a strong recommendation. Factors that favor a decision to
secondary prevention trials will achieve an LDL-C level reduce LDL-C levels to ⬍70 mg/dL are those that place
⬍100 mg/dL in little more than half of high-risk patients patients in the category of very high risk. Among these factors
(4,26 –28). To attain an LDL-C ⬍100 mg/dL in the remaining are the presence of established CVD plus (1) multiple major
patients, either the statin dose must be increased or a second risk factors (especially diabetes), (2) severe and poorly
LDL-lowering drug must be added to therapy. Thus, in ATP controlled risk factors (especially continued cigarette smok-
III, the LDL-C goal of ⬍100 mg/dL was considered to be not ing), (3) multiple risk factors of the metabolic syndrome
only the limit of efficacy supported by available clinical trial (especially high triglycerides ⱖ200 mg/dL plus non-HDL-C
data but also the practical limit that could be achieved in most ⱖ130 mg/dL with low HDL-C [⬍40 mg/dL]), and (4) on the
high-risk patients with standard therapy as informed by basis of PROVE IT, patients with acute coronary syndromes.
clinical trials. To avoid any misunderstanding about cholesterol manage-
ment in general, it must be emphasized that the optional goal
Rationale for Optional Very Low LDL-C of ⬍70 mg/dL does not apply to individuals who are not high
Goal (<70 mg/dL) risk.
A question raised by HPS and PROVE IT is whether an
LDL-C goal of ⬍100 mg/dL is sufficiently low in high-risk Potential Side Effects of Very Low LDL Cholesterol
patients who already have a low LDL-C level at baseline. In In the past, concern has been raised about potential dangers of
HPS, patients whose LDL-C levels at baseline were ⬍116 reducing LDL to very low levels. Some epidemiological
mg/dL, and even the subgroup with LDL-C concentrations studies (31–33) suggest that very low serum cholesterol
⬍100 mg/dL, exhibited significant risk reduction when statin levels are associated with an increase in total mortality. In
therapy was introduced. In PROVE IT, intensive LDL-C– particular, an association with cerebral hemorrhage has been
lowering therapy with high-dose statin (atorvastatin) reduced reported. In these studies, a causal link between low choles-
major cardiovascular events in only 2 years as compared with terol levels and morbidity or mortality has not been estab-
standard-dose statin (pravastatin 40 mg). Pravastatin 40 mg lished. Some investigators attribute the association to con-
reduced the median LDL-C from 106 mg/dL to 95 mg/dL, founding factors. In recent clinical trials with statin therapy,
which achieved the ATP III goal of ⬍100 mg/dL; atorvastatin no significant side effects from LDL lowering per se have
80 mg lowered LDL-C to a median of 62 mg/dL. Thus, on the been identified. For these reasons, the decision to achieve
basis of both HPS and PROVE IT, an LDL-C level of 100 very low LDL levels in very high-risk patients should be
726 Grundy et al. JACC Vol. 44, No. 3, 2004
Recent Clinical Trials and NCEP ATP III August 4, 2004:720–32

TABLE 1. Doses of Currently Available Statins Required to likewise be attained by combining lower doses of statins with
Attain an Approximate 30% to 40% Reduction of LDL-C Levels other drugs or products (eg, bile acid sequestrants, nicotinic
(Standard Doses)* acid, ezetimibe, plant stanols/sterols). Because of the avail-
Drug Dose, mg/d LDL Reduction, % ability of a variety of relatively safe LDL-lowering options,
when ATP III indicates that drug therapy should be consid-
Atorvastatin 10† 39
ered, it is reasonable to employ doses adequate to achieve a
Lovastatin 40† 31
reduction in risk for major coronary events of 30% to 40%.
Pravastatin 40† 34 To use minimal drug therapy just to produce a small LDL
Simvastatin 20–40† 35–41 reduction that will barely attain the LDL-C goal would not be
Fluvastatin 40–80 25–35 a prudent use of LDL-lowering drugs. These comments must
Rosuvastatin 5–10‡ 39–45 not be taken to mean that NCEP is recommending a 30% to
*Estimated LDL reductions were obtained from US Food and Drug Admin- 40% reduction of LDL-C levels as a goal of therapy. The
istration package inserts for each drug. comments simply recognize that if drug therapy is a compo-
†All of these are available at doses up to 80 mg. For every doubling of the nent of cholesterol management for a given patient, it is
dose above standard dose, an approximate 6% decrease in LDL-C level can be prudent to employ doses that will achieve at least a moderate
obtained (45). risk reduction.
‡For rosuvastatin, doses available up to 40 mg; the efficacy for 5 mg is
estimated by subtracting 6% from the Food and Drug Administration–reported
Because of the success of statin trials, some investigators
efficacy at 10 mg (45). have suggested that guidelines can be simplified by merely
recommending that high-risk patients be treated with the
doses of statins used in clinical trials. In the view of NCEP,
based on evidence of benefit and recognition that there this suggestion represents an oversimplification that will lead
appears to be only a remote possibility of side effects from to undertreatment of many patients. It does not take advan-
LDL lowering per se. tage of the strong database supporting the log-linear relation-
ship between LDL levels and CHD risk (Figure). As shown in
Limitations in Efficacy of LDL-Lowering Therapy HPS, if a high-risk patient has a relatively low LDL concen-
In spite of growing evidence for benefit of reducing LDL-C tration at baseline, a standard dose of statin may achieve the
levels to ⬍70 mg/dL in very high-risk patients, many such minimal LDL-C goal of ⬍100 mg/dL or even the more
patients may not be able to achieve such low levels with stringent optional goal of ⬍70 mg/dL. For persons with
currently available drugs. This will be the case particularly higher LDL levels at baseline, standard doses of statins may
when baseline LDL-C levels are relatively high. For example, fail to achieve an LDL-C level ⬍100 mg/dL and thus may not
even with high-dose statins (34) or LDL-lowering drug achieve the full potential of benefit from LDL lowering. As
combinations (35,36), LDL-C reductions ⬎50% often cannot the number of LDL-lowering options increases, the initiation
be achieved. Thus, when baseline LDL-C is ⬎150 mg/dL, it of more intensive therapies becomes feasible. NCEP recom-
may not be possible to achieve an LDL-C ⬍70 mg/dL in very mends that such therapies be employed within the bounds of
high-risk patients. safety and tolerability to at least achieve an LDL-C level of
⬍100 mg/dL.
Relation of Percentage Reduction in LDL to
CHD Risk: Implications for Therapy
ATP III recommendations on therapy placed higher priority Implications of HPS and PROVE IT for
on reaching the LDL-C goals than on achieving a given Clinical Management of Elevated LDL-C in
percentage lowering of LDL-C levels. ATP III guidelines also High-Risk Patients
identified characteristics of persons in whom cholesterol- HPS in general supports ATP III guidelines for high-risk
lowering drugs should be considered. The guidelines, how- patients. The introduction of the concept of CHD risk equiva-
ever, were not explicit on how much LDL-C lowering should lents in ATP III expanded the definition of high risk beyond
be sought from drug therapy beyond achieving the LDL-C established CHD to include other types of high-risk patients.
goal. Recent clinical trials nonetheless have documented how Because the LDL-C treatment goal for all of these categories is
much reduction in relative risk for major coronary events can a level ⬍100 mg/dL, the majority of high-risk patients will
be achieved from a given lowering of LDL-C (4 –7,26 – require intensive LDL-lowering therapy. By coincidence or
28,37,38). They indicate that for every 1% reduction in design, HPS included several different types of high-risk patients
LDL-C levels, relative risk for major CHD events is reduced that would qualify as CHD risk equivalents according to ATP
by approximately 1%. HPS data suggest that this relationship III. The benefit of LDL-lowering therapy in such high-risk
holds for LDL-C levels even below 100 mg/dL (Figure). patients was amply demonstrated by HPS. Importantly, HPS
Currently available statins at doses typically used in these provides support for the use of intensive LDL-C lowering in
trials will lower LDL-C levels by 30% to 40%, which most high-risk patients. The implications of HPS for different
translates into a similar percentage reduction in CHD risk levels of LDL-C in high-risk patients thus can be considered.
over a 5-year period. In the present document, the statin doses
that produce such reductions are called standard doses. Table Baseline LDL-C Levels >130 mg/dL
1 lists these standard doses for currently available statin For high-risk persons, ATP III recommended that LDL-
drugs. Similar reductions in LDL-C of 30% to 40% can lowering drugs begin simultaneously with dietary therapy
JACC Vol. 44, No. 3, 2004 Grundy et al. 727
August 4, 2004:720–32 Recent Clinical Trials and NCEP ATP III

when LDL-C is ⱖ130 mg/dL. HPS supports this recommen- mg/dL, eg, to reduce LDL-C to the range of ⬍70 mg/dL, is a
dation; those HPS subjects with higher LDL-C levels had the reasonable therapeutic decision on the basis of clinical
greatest reduction in absolute risk from statin therapy. As judgment that the patient is still at very high absolute risk for
shown in several clinical trials (26 –28), including HPS, future CVD events. This therapeutic strategy is supported by
however, when LDL-C levels are well above 130 mg/dL, eg, the results of HPS and PROVE IT. For LDL-C ⬍100 mg/dL,
ⱖ160 mg/dL, standard doses of statins may not be sufficient other lipid-lowering drugs (eg, fibrates, nicotinic acid) can be
to achieve the goal of ⬍100 mg/dL. When they do not, the considered for patients with elevated triglycerides and/or low
dose of statin may have to be increased or a second agent (eg, HDL-C; these drugs can be used, either as alternatives to statin
ezetimibe, bile acid sequestrant, or nicotinic acid) may be therapy, as shown by the Veterans Affairs High-Density Li-
needed. Alternatively, a maximizing of dietary therapy (in- poprotein Cholesterol Intervention Trial (VA-HIT) (40), or in
cluding the use of plant stanols/sterols) combined with a combination with statins.
standard dose of statin may be sufficient to attain the ATP III
goal in some patients. A recent report indicates that maximal On-Treatment LDL-C <100 mg/dL
dietary therapy can achieve LDL-C reductions of up to 25% Again, ATP III did not recommend further LDL-lowering
to 30% (39). Combined with standard doses of statins, such therapy for this group. A log-linear relationship between
dietary therapy should lower LDL-C levels by well above LDL-C level and CHD risk implies that further reduction in
40%, which often will achieve the recommended target of risk could be achieved by still more LDL lowering (Figure).
therapy. HPS results are consistent with this possibility, and so are
those of PROVE IT, but these results cannot be considered
definitive. Several clinical trials are currently underway (see
Baseline LDL-C Levels of 100 to 129 mg/dL
Waters et al (30)) in which standard-dose and high-dose statin
By setting an LDL-C goal of ⬍100 mg/dL, ATP III favored
therapy are being compared. Moreover, to achieve the LDL-C
institution of LDL-lowering therapy in this LDL-C range.
goal of ⬍100 mg/dL, many patients may already have been
Still, for patients having low HDL levels as the predominant
treated with either high doses of statins or combined drug
lipoprotein abnormality, fibrates or nicotinic acid were ac-
therapy. In such patients, achieving a yet lower LDL goal (eg,
knowledged as alternatives to statin therapy. Recent clinical
⬍70 mg/dL) will not be a practical option. For those patients
trials with fibrate therapy are consistent with this option (40).
who attain an LDL-C ⬍100 mg/dL on standard doses of
HPS results, on the other hand, reinforce the ATP III–preferred
statins, physicians can consider intensifying LDL-C reduc-
option, ie, institution of LDL-lowering drug therapy. The HPS
tion. Intensified therapy might be reserved for those patients
finding of a substantial benefit from use of a standard dose of
deemed to be at very high risk. PROVE IT reported 2-year
statin further implies that for those with baseline LDL-C close to
benefit from intensified LDL lowering in patients with acute
100 mg/dL, therapy should be intensive enough to achieve a
coronary syndromes, and it will be important to confirm these
30% to 40% reduction in LDL-C levels, and not merely enough
results through several other ongoing clinical trials (see
statin to attain an LDL-C level just below 100 mg/dL. A small
Waters et al (30)) of similar design before making global
lowering of LDL-C just to achieve the goal will not yield much
recommendations for high-risk patients with on-treatment
additional risk reduction. Standard doses of statins, in contrast,
LDL-C ⬍100 mg/dL.
are sufficient to attain a substantial risk reduction. If nicotinic
acid or fibrates are considered an option for this LDL-C range,
Patients With Acute Coronary Syndromes
it may be preferable to use them in combination with an These patients are at very high risk for suffering recurrent
LDL-lowering drug and not as a sole agent. coronary events in the near term. The Myocardial Ischemia
Reduction with Aggressive Cholesterol Lowering (MIRACL)
Baseline LDL-C Levels <100 mg/dL trial (41) previously suggested that intensive LDL-lowering
ATP III did not recommend institution of LDL-lowering therapy would reduce risk for recurrent cardiovascular events
therapy in high-risk patients when the serum LDL-C is ⬍100 in the first 18 months after acute coronary syndromes.
mg/dL. HPS, however, found that such patients had a signif- PROVE IT greatly strengthens the evidence for benefit of
icant lowering of risk for CVD events when they were treated intensive LDL lowering in the first 2 years after acute
with a standard dose of statin. On the basis of HPS, some coronary syndromes. For this reason, intensive therapy
authorities recommend the use of statin therapy in virtually should be considered for all patients admitted to the hospital
all high-risk patients whose LDL-C is ⬍100 mg/dL. Indeed, for acute coronary syndromes. A strong case is made by
further risk reduction through LDL lowering in patients with PROVE IT for achieving the optional LDL-C goal of ⬍70
high baseline risk is consistent with the log-linear relationship mg/dL. Choice of drug and dosage should be guided in part
between LDL-C levels and CHD risk shown in the Figure. by measurement of LDL-C within 24 hours of admission to
However, a global recommendation to lower LDL-C in the hospital. Modification of therapy can be made at
high-risk patients with LDL-C ⬍100 mg/dL cannot be based follow-up if necessary to achieve the desired LDL-C level. If
on HPS alone in light of the limitations discussed before the baseline in-hospital LDL-C is relatively low, even an
(page 229). Ongoing clinical trials may provide additional LDL-C level of ⬍70 mg/dL may be achieved by a standard
support for recommending an LDL-C goal well below 100 dose of statin. If the baseline LDL-C level is higher, a high
mg/dL. In the meantime, initiation of an LDL-lowering drug dose of statin or the combination of a standard dose of statin
in high-risk patients when baseline serum LDL-C is ⬍100 with ezetimibe, bile acid sequestrant, or nicotinic acid may be
728 Grundy et al. JACC Vol. 44, No. 3, 2004
Recent Clinical Trials and NCEP ATP III August 4, 2004:720–32

required. In choice of therapy, consideration should be given diabetes is considered to be at lower risk, an LDL-lowering
to safety of the regimen for the individual patient as well as drug might not be started if the LDL-C level is ⬍130 mg/dL.
to efficacy of treatment. Maximal TLC clearly is indicated, but clinical judgment must
be exercised with regard to when to initiate an LDL-lowering
Implications of HPS Results for Patients drug.
With Diabetes
ATP III identified diabetes as a high-risk condition. This Implications of HPS, PROSPER, and ASCOT
designation was based on evidence that the majority of for Cholesterol Management in Older Persons
patients with diabetes in higher-risk populations have a ATP III counseled that, on the basis of considerations of age
relatively high 10-year risk for developing CVD. In addition, alone, older persons should not be denied the benefits of
the onset of CVD in patients with diabetes carries a poor LDL-lowering therapy accorded to other age groups. Al-
prognosis, both at the time of an acute CVD event and in the though several epidemiological studies found that elevated
post-event period. Moreover, clinical trials (42,43) before cholesterol levels confer a smaller relative risk in older
HPS provided moderately strong evidence that LDL-lowering compared with younger persons, the absolute risk attributable
therapy is efficacious in patients with diabetes. HPS investi- to increased cholesterol levels remains high. Moreover, sub-
gators recently carried out and reported a detailed analysis of group analysis of several previous trials with statins strongly
their results in patients with diabetes (11). The results of this suggested that LDL-lowering therapy significantly reduces
analysis can be considered in relation to ATP III risk for CHD in older persons. HPS and PROSPER results
recommendations. add support for benefit of LDL-lowering therapy in older
persons. The implications for 2 groups of older persons can
Diabetes Plus CVD be examined briefly.
In HPS, patients who had both diabetes and CVD were at
very high risk for future CVD events. In terms of absolute Older Persons With Established CVD
risk reduction, this category of patient obtained the greatest HPS explicitly documented risk reduction with statin therapy
benefit from statin therapy. Therefore, patients with the in older persons (65 to 80 years) at high risk. Absolute risk
combination of diabetes and CVD deserve intensive lipid- reduction was just as great in this group as in other high-risk
lowering therapy. On the basis of HPS, the presence of this groups. Older persons tolerated statin therapy well. Although
combination appears to support initiation of statin therapy PROSPER had fewer older persons with established CVD,
regardless of baseline LDL-C levels. For patients with dia- and they were treated for a shorter time than in HPS, a strong
betes plus CVD, it is reasonable to attempt to achieve a very trend toward reduction in CHD was noted. The results of HPS
low LDL-C level (eg, ⬍70 mg/dL). and PROSPER, taken together with the findings of other
statin trials, provide a strong justification for intensive LDL-
Diabetes Without CVD lowering therapy in older persons with established CVD.
ATP III indicated that most patients with diabetes are at high
risk even in the absence of established CHD. Most patients Older Persons at High Risk Without
with hyperglycemia have type 2 diabetes, are older, and have Established CVD
multiple risk factors. Epidemiological studies and clinical Absolute risk rises with age because of progressive accumu-
trials demonstrate that in higher-risk populations these pa- lation of coronary atherosclerosis (1,2). Women are at lower
tients have a risk for CVD events approximately equal to that risk, but if they have multiple risk factors, they too are at
of nondiabetic patients with established CVD. HPS data relatively high risk. Older patients with diabetes certainly
found both a high risk in this group and benefit from must be considered to be at high risk. Unfortunately, risk
LDL-lowering therapy, supporting the LDL-C goal of ⬍100 assessment in older persons is not highly reliable. Other tests
mg/dL. On the other hand, in those diabetic patients without in older persons without clinical CVD hold promise for
CVD who had an LDL-C at baseline of ⬍116 mg/dL, risk improving risk estimates, but so far, additional testing has not
reduction accompanying statin therapy was only marginally been integrated into quantitative risk assessment. Therefore,
significant for first coronary event. Thus, whether to start an beyond use of Framingham risk scoring in older persons,
LDL-lowering drug when LDL-C is ⬍100 mg/dL in this clinical judgment is required as to when to initiate intensive
category of patient must be left to clinical judgment. LDL-lowering therapy in older persons without CVD. Effi-
As noted in ATP III, not all patients with clinical diabetes cacy alone is not the key issue in this group. A host of factors
have a 10-year risk ⬎20%. Many of those who do not must be weighed, including efficacy, safety, tolerability, and
nonetheless deserve to be classified as high risk because of patient preference, in this age group. The results of both
poor prognosis once CHD becomes manifest, as mentioned PROSPER and ASCOT support the efficacy of statin therapy
before. On the other hand, a portion of patients with diabetes in older, high-risk persons without established CVD.
can be considered to be at only moderately high risk because
of young age or lack of other risk factors. Such patients were Implications of the ASCOT-LLA and
not studied in HPS. For the category of moderately high risk ALLHAT-LLT Trials for Patients at
(10-year risk 10% to 20%), ATP III guidelines favored Moderately High Risk
institution of LDL-lowering drugs along with dietary therapy ATP III identified a specific risk category that includes
when LDL-C levels are ⱖ130 mg/dL. Thus, if a patient with people with 2⫹ risk factors and a 10-year risk of 10% to 20%
JACC Vol. 44, No. 3, 2004 Grundy et al. 729
August 4, 2004:720–32 Recent Clinical Trials and NCEP ATP III

TABLE 2. ATP III LDL-C Goals and Cutpoints for TLC and Drug Therapy in Different Risk Categories and Proposed Modifications
Based on Recent Clinical Trial Evidence
Risk Category LDL-C Goal Initiate TLC Consider Drug Therapy**
High risk: CHD* or CHD risk equivalents† ⬍100 mg/dL ⱖ100 mg/dL# ⱖ100 mg/dL††
(10-year risk ⬎20%) (optional goal: ⬍70 mg/dL)储 (⬍100 mg/dL: consider drug options)**
Moderately high risk: 2⫹ risk factors‡ ⬍130 mg/dL¶ ⱖ130 mg/dL# ⱖ130 mg/dL
(10-year risk 10% to 20%)§§ (100–129 mg/dL; consider drug options)‡‡
Moderate risk: 2⫹ risk factors‡ (10-year ⬍130 mg/dL ⱖ130 mg/dL ⱖ160 mg/dL
risk ⬍10%)§§
Lower risk: 0–1 risk factor§ ⬍160 mg/dL ⱖ160 mg/dL ⱖ190 mg/dL
(160–189 mg/dL: LDL-lowering drug optional)
*CHD includes history of myocardial infarction, unstable angina, stable angina, coronary artery procedures (angioplasty or bypass surgery), or evidence of clinically
significant myocardial ischemia.
†CHD risk equivalents include clinical manifestations of noncoronary forms of atherosclerotic disease (peripheral arterial disease, abdominal aortic aneurysm, and
carotid artery disease 关transient ischemic attacks or stroke of carotid origin or ⬎50% obstruction of a carotid artery兴), diabetes, and 2⫹ risk factors with 10-year
risk for hard CHD ⬎20%.
‡Risk factors include cigarette smoking, hypertension (BP ⱖ140/90 mm Hg or on antihypertensive medication), low HDL cholesterol (⬍40 mg/dL), family history
of premature CHD (CHD in male first-degree relative ⬍55 years of age; CHD in female first-degree relative ⬍65 years of age), and age (men ⱖ45 years; women
ⱖ55 years).
§§Electronic 10-year risk calculators are available at www.nhlbi.nih.gov/guidelines/cholesterol.
§Almost all people with zero or 1 risk factor have a 10-year risk ⬍10%, and 10-year risk assessment in people with zero or 1 risk factor is thus not necessary.
储Very high risk favors the optional LDL-C goal of ⬍70 mg/dL, and in patients with high triglycerides, non-HDL-C ⬍100 mg/dL.
¶Optional LDL-C goal ⬍100 mg/dL.
#Any person at high risk or moderately high risk who has lifestyle-related risk factors (eg, obesity, physical inactivity, elevated triglyceride, low HDL-C, or metabolic
syndrome) is a candidate for therapeutic lifestyle changes to modify these risk factors regardless of LDL-C level.
**When LDL-lowering drug therapy is employed, it is advised that intensity of therapy be sufficient to achieve at least a 30% to 40% reduction in LDL-C levels.
††If baseline LDL-C is ⬍100 mg/dL, institution of an LDL-lowering drug is a therapeutic option on the basis of available clinical trial results. If a high-risk person
has high triglycerides or low HDL-C, combining a fibrate or nicotinic acid with an LDL-lowering drug can be considered.
‡‡For moderately high-risk persons, when LDL-C level is 100 to 129 mg/dL, at baseline or on lifestyle therapy, initiation of an LDL-lowering drug to achieve an
LDL-C level ⬍100 mg/dL is a therapeutic option on the basis of available clinical trial results.

(moderately high risk). Individuals in this category were ALLHAT-LLT recruited a heterogeneous group of sub-
considered to be candidates for LDL-lowering drugs if their jects that on average appear to fall into the moderately
serum LDL-C after TLC is ⱖ130 mg/dL. The LDL-C goal for high-risk category. The results in ALLHAT were disappoint-
these persons was set at a level of ⬍130 mg/dL. If LDL- ing because of the small difference in cholesterol levels
lowering drugs are employed to achieve the LDL-C goal between usual-care and statin-therapy groups. It should be
recommended by ATP III, presumably the dose of drug noted, however, that a significant reduction in risk for major
should be sufficient to reduce LDL-C levels by 30% to 40%. cardiovascular events was obtained in the African-American
ATP III did not recommend LDL-lowering therapy in subgroup treated with pravastatin; this finding supports the
moderately high-risk patients in whom serum LDL-C is ATP III recommendation that goals of LDL-lowering therapy
⬍130 mg/dL. However, a significant portion of the subjects should not be modified on the basis of ethnicity.
in the ASCOT study, who had LDL-C ⬍130 mg/dL and were For people in lower risk categories (2⫹ risk factors and
at moderately high risk by ATP III criteria, had a significant 10-year risk ⬍10%, or 0 to 1 risk factor), the results of recent
lowering of risk for CVD when they were treated with a clinical trials do not modify the goals and cutpoints of
standard dose of a statin. Thus, ASCOT supports use of an therapy.
LDL-lowering drug in persons with a 10-year risk of 10% to
20% and LDL-C level of 100 to 129 mg/dL, at baseline or on
lifestyle changes, to achieve an LDL-C level ⬍100 mg/dL, as Summary of Implications of Recent Clinical
a therapeutic option on the basis of clinical judgment of the Trials for ATP III Treatment Algorithm
patient’s absolute risk and potential benefit of an LDL- From the evidence of previous statin trials, the ATP III panel
lowering drug. Initiation of TLC also is recommended. was able to expand both the scope and intensity of LDL-
Factors that might favor use of an LDL-lowering drug in this lowering therapy for higher-risk individuals beyond that
category include advancing age, more than 2 risk factors, recommended in ATP II. The number of Americans for
severe risk factors (eg, continued cigarette smoking, a whom LDL-lowering drugs are considered was significantly
strongly positive family history of premature atherosclerotic increased by ATP III. Recent statin trials have provided new
CVD), high triglycerides (ⱖ 200 mg/dL) plus elevated information on benefits of LDL-lowering therapy applied to
non-HDL-C (ⱖ160 mg/dL), low HDL-C (⬍40 mg/dL), the persons in categories in which ATP III could not make
metabolic syndrome, and/or the presence of emerging risk definitive recommendations about drug therapy. In general,
factors (eg, serum high-sensitivity C-reactive protein ⬎3 these new trials have strongly reinforced ATP III recommen-
mg/L (2,44) or coronary calcium ⬎75th percentile for a dations. In particular, they support ATP III recommendations
person’s age and sex (2)). for the benefit of LDL-lowering therapy for patients with
730 Grundy et al. JACC Vol. 44, No. 3, 2004
Recent Clinical Trials and NCEP ATP III August 4, 2004:720–32

TABLE 3. Recommendations for Modifications to Footnote the ATP III Treatment Algorithm for LDL-C
● Therapeutic lifestyle changes (TLC) remain an essential modality in clinical management. TLC has the potential to reduce cardiovascular risk through several
mechanisms beyond LDL lowering.
● In high-risk persons, the recommended LDL-C goal is ⬍100 mg/dL.
⫺ An LDL-C goal of ⬍70 mg/dL is a therapeutic option on the basis of available clinical trial evidence, especially for patients at very high risk.
⫺ If LDL-C is ⱖ100 mg/dL, an LDL-lowering drug is indicated simultaneously with lifestyle changes.
⫺ If baseline LDL-C is ⬍100 mg/dL, institution of an LDL-lowering drug to achieve an LDL-C level ⬍70 mg/dL is a therapeutic option on the basis of
available clinical trial evidence.
⫺If a high-risk person has high triglycerides or low HDL-C, consideration can be given to combining a fibrate or nicotinic acid with an LDL-lowering drug.
When triglycerides are ⱖ200 mg/dL, non-HDL-C is a secondary target of therapy, with a goal 30 mg/dL higher than the identified LDL-C goal.
● For moderately high-risk persons (2⫹ risk factors and 10-year risk 10% to 20%), the recommended LDL-C goal is ⬍130 mg/dL; an LDL-C goal ⬍100
mg/dL is a therapeutic option on the basis of available clinical trial evidence. When LDL-C level is 100 to 129 mg/dL, at baseline or on lifestyle therapy,
initiation of an LDL-lowering drug to achieve an LDL-C level ⬍100 mg/dL is a therapeutic option on the basis of available clinical trial evidence.
● Any person at high risk or moderately high risk who has lifestyle-related risk factors (eg, obesity, physical inactivity, elevated triglyceride, low HDL-C, or
metabolic syndrome) is a candidate for TLC to modify these risk factors regardless of LDL-C level.
● When LDL-lowering drug therapy is employed in high-risk or moderately high-risk persons, it is advised that intensity of therapy be sufficient to achieve at
least a 30% to 40% reduction in LDL-C levels.
● For people in lower-risk categories, recent clinical trials do not modify the goals and cutpoints of therapy.

diabetes and in older persons. Moreover, they provide new risk patients who have elevated triglycerides or low HDL-C
information on the efficacy of risk reduction in high-risk levels, addition of a fibrate or nicotinic acid to LDL-lowering
persons with relatively low LDL-C levels. Although the full therapy can be considered.
benefit of LDL-C reduction in higher-risk patients with low For patients at moderately high risk (10-year risk 10% to
or very low LDL-C levels is still under investigation, the 20%), the LDL-C goal remains ⬍130 mg/dL. However, a
recent results open the door to use of cholesterol-lowering goal of ⬍100 mg/dL represents a therapeutic option on the
drugs in such patients with very high absolute risk who are basis of evidence of efficacy in risk reduction from primary-
most likely to benefit from added therapy. prevention trials. TLC should be initiated in all such persons
Table 2 shows the ATP III goals and cutpoints and whose LDL-C level is ⱖ130 mg/dL. Again, any person at
proposed modifications in the treatment algorithm for LDL moderately high risk who has lifestyle-related risk factors (eg,
cholesterol based on evidence from recent clinical trials. obesity, physical inactivity, elevated triglycerides, low
Essential modifications are highlighted in footnotes to Table HDL-C, or metabolic syndrome) is a candidate for TLC to
2 and are summarized in Table 3. Several modifications offer modify these risk factors regardless of LDL-C level. If the
therapeutic options with regard to LDL-C goals lower than LDL-C concentration is ⱖ130 mg/dL after TLC, consider-
those in ATP III and choice of therapies. Recent clinical trials ation should be given to initiating an LDL-lowering drug, to
provide greater rationale for more intensive LDL-lowering achieve and sustain the LDL-C goal of ⬍130 mg/dL. For
therapy, but they do not resolve all issues surrounding very LDL-C levels of 100 to 129 mg/dL at baseline or on lifestyle
low LDL levels. At these levels, physicians must ultimately therapy, initiation of an LDL-lowering drug to achieve an
rely on clinical judgment to weigh patient risk and the LDL-C level ⬍100 mg/dL is a therapeutic option on the basis
efficacy, safety, and cost of different therapies. These issues of clinical trial evidence of additional efficacy.
can be discussed in the following context. When initiating LDL-lowering therapy in a person at high
For high-risk patients, the recommended LDL-C treatment risk or moderately high risk, the efficacy of therapeutic
goal remains at ⬍100 mg/dL. However, a target of ⬍70 lifestyle change both to lower LDL-C levels and to reduce
mg/dL represents a therapeutic option, ie, a reasonable risk through other mechanisms must not be overlooked.
clinical strategy, for persons considered to be at very high Lifestyle change must be an integral part of risk reduction
risk, on the basis of emerging clinical trial data. TLC is therapy. When an LDL-lowering drug is employed in a
recommended in high-risk patients whenever the LDL-C person at high risk or moderately high risk, a reduction in
level is ⱖ100 mg/dL. Furthermore, any person at high risk LDL-C levels of at least 30% to 40% beyond dietary therapy
who has lifestyle-related risk factors (eg, obesity, physical should be achieved if feasible. For people in lower risk
inactivity, elevated triglycerides, low HDL-C, or metabolic categories, there are no proposed changes to the treatment
syndrome) is a candidate for TLC to modify these risk factors goals and cutpoints (45).
regardless of LDL-C level. As before, whenever the baseline
LDL-C concentration is ⱖ130 mg/dL, simultaneous initiation References
of an LDL-lowering drug and dietary therapy is recom- 1. Expert Panel on Detection, Evaluation, and Treatment of High Blood
mended. If LDL-C is 100 to 129 mg/dL, the same now holds. Cholesterol in Adults. Executive Summary of the Third Report of the
If baseline LDL-C is ⬍100 mg/dL and the patient is consid- National Cholesterol Education Program (NCEP) Expert Panel on
Detection, Evaluation, and Treatment of High Blood Cholesterol in
ered to be at very high risk, initiation of an LDL-lowering
Adults (Adult Treatment Panel III). JAMA 2001;285:2486 –97.
drug to achieve an LDL-C level of ⬍70 mg/dL is a thera- 2. National Cholesterol Education Program (NCEP) Expert Panel on
peutic option that has clinical trial support. For those high- Detection, Evaluation, and Treatment of High Blood Cholesterol in
JACC Vol. 44, No. 3, 2004 Grundy et al. 731
August 4, 2004:720–32 Recent Clinical Trials and NCEP ATP III

Adults (Adult Treatment Panel III). Third Report of the National Cho- statin and simvastatin (the ADvicor Versus Other Cholesterol-Modulating
lesterol Education Program (NCEP) Expert Panel on Detection, Eval- Agents Trial Evaluation [ADVOCATE]). Am J Cardiol 2003;91:667–72.
uation, and Treatment of High Blood Cholesterol in Adults (Adult 23. Law MR, Wald NJ, Thompson SG. By how much and how quickly does
Treatment Panel III) final report. Circulation 2002;106:3143–21. reduction in serum cholesterol concentration lower risk of ischaemic heart
3. National Cholesterol Education Program. Second Report of the Expert disease? BMJ 1994;308:367–72.
Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol 24. Law MR, Wald NJ. An ecological study of serum cholesterol and
in Adults (Adult Treatment Panel II). Circulation 1994;89:1333– 445. ischaemic heart disease between 1950 and 1990. Eur J Clin Nutr 1994;
4. Heart Protection Study Collaborative Group. MRC/BHF Heart Pro- 48:305–25.
tection Study of cholesterol lowering with simvastatin in 20,536 25. Law MR, Wald NJ, Rudnicka AR. Quantifying effect of statins on low
high-risk individuals: a randomised placebo-controlled trial. Lancet density lipoprotein cholesterol, ischaemic heart disease, and stroke: sys-
2002;360:7–22. tematic review and meta-analysis. BMJ 2003;326:1423.
5. Shepherd J, Blauw GJ, Murphy MB, et al., PROSPER study group. 26. Randomised trial of cholesterol lowering in 4444 patients with coronary
Pravastatin in elderly individuals at risk of vascular disease (PROSPER): heart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet
a randomised controlled trial. PROspective Study of Pravastatin in the 1994;344:1383–9.
Elderly at Risk. Lancet 2002;360:1623–30. 27. Sacks FM, Pfeffer MA, Moye LA, et al. The effect of pravastatin on
6. ALLHAT Officers and Coordinators for the ALLHAT Collaborative
coronary events after myocardial infarction in patients with average
Research Group. The Antihypertensive and Lipid-Lowering Treatment to
cholesterol levels: Cholesterol and Recurrent Events Trial investigators.
Prevent Heart Attack Trial. Major outcomes in moderately hypercholes-
N Engl J Med 1996;335:1001–9.
terolemic, hypertensive patients randomized to pravastatin vs usual care:
28. The Long-Term Intervention with Pravastatin in Ischaemic Disease
the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart
(LIPID) Study Group. Prevention of cardiovascular events and death with
Attack Trial (ALLHAT-LLT). JAMA 2002;288:2998 –3007.
pravastatin in patients with coronary heart disease and a broad range of
7. Sever PS, Dahlof B, Poulter NR, et al., ASCOT investigators. Prevention
initial cholesterol levels. N Engl J Med 1998;339:1349 –57.
of coronary and stroke events with atorvastatin in hypertensive patients
who have average or lower-than-average cholesterol concentrations, in 29. Nissen SE, Tuzcu EM, Schoenhagen P, et al., REVERSAL Investigators.
the Anglo-Scandinavian Cardiac Outcomes Trial–Lipid Lowering Arm Effect of intensive compared with moderate lipid-lowering therapy on
(ASCOT-LLA): a multicentre randomised controlled trial. Lancet 2003; progression of coronary atherosclerosis: a randomized controlled trial.
361:1149 –58. JAMA 2004;291:1071– 80.
8. Cannon CP, Braunwald E, McCabe CH, et al., Pravastatin or Atorvastatin 30. Waters DD, Guyton JR, Herrington DM, McGowan MP, Wenger NK,
Evaluation and Infection Therapy-Thrombolysis in Myocardial Infarction Shear C, TNT Steering Committee Members and Investigators. Treating
22 Investigators. Intensive versus moderate lipid lowering with statins to New Targets (TNT) Study: does lowering low-density lipoprotein
after acute coronary syndromes. N Engl J Med 2004;350:1495–504. cholesterol levels below currently recommended guidelines yield incre-
9. McNamara JR, Cole TG, Contois JH, Ferguson CA, Ordovas JM, mental clinical benefit? Am J Cardiol 2004;93:154 – 8.
Schaefer EJ. Immunoseparation method for measuring low-density 31. Stemmermann GN, Chyou PH, Kagan A, Nomura AM, Yano K. Serum
lipoprotein cholesterol directly from serum evaluated. Clin Chem 1995; cholesterol and mortality among Japanese-American men: the Honolulu
41:232– 40. (Hawaii) Heart Program. Arch Intern Med 1991;151:969 –72.
10. Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concen- 32. Neaton JD, Blackburn H, Jacobs D, et al. Serum cholesterol level and
tration of low-density lipoprotein cholesterol in plasma, without use of the mortality findings for men screened in the Multiple Risk Factor Inter-
preparative ultracentrifuge. Clin Chem 1972;18:499 –502. vention Trial: Multiple Risk Factor Intervention Trial Research Group.
11. Collins R, Armitage J, Parish S, Sleigh P, Peto R, Heart Protection Study Arch Intern Med 1992;152:1490 –500.
Collaborative Group. MRC/BHF Heart Protection Study of cholesterol- 33. Iso H, Naito Y, Kitamura A, et al. Serum total cholesterol and mortality
lowering with simvastatin in 5963 people with diabetes: a randomised in a Japanese population. J Clin Epidemiol 1994;47:961–9.
placebo-controlled trial. Lancet 2003;361:2005–16. 34. Jones PH, Davidson MH, Stein EA, et al., STELLAR Study Group.
12. Sacks FM, Tonkin AM, Shepherd J, et al. Effect of pravastatin on Comparison of the efficacy and safety of rosuvastatin versus atorvastatin,
coronary disease events in subgroups defined by coronary risk factors: the simvastatin, and pravastatin across doses (STELLAR* Trial). Am J
Prospective Pravastatin Pooling Project. Circulation 2000;102:1893–900. Cardiol 2003;92:152– 60.
13. Dean BB, Borenstein JE, Henning JM, Knight K, Bairey Merz CN. Can 35. Davidson MH, McGarry T, Bettis R, et al. Ezetimibe coadministered with
change in HDL-cholesterol reduce cardiovascular risk? Am Heart J 2004; simvastatin in patients with primary hypercholesterolemia. J Am Coll
147:966 –76. Cardiol 2002;40:2125–34.
14. Rubins HB. Triglycerides and coronary heart disease: implications of 36. Ballantyne CM, Houri J, Notarbartolo A, et al., Ezetimibe Study Group.
recent clinical trials. J Cardiovasc Risk 2000;7:339 – 45. Effect of ezetimibe coadministered with atorvastatin in 628 patients with
15. Robins SJ, Rubins HB, Faas FH, et al., Veterans Affairs HDL Inter- primary hypercholesterolemia: a prospective, randomized, double-blind
vention Trial (VA-HIT). Insulin resistance and cardiovascular events with trial. Circulation 2003;107:2409 –15.
low HDL cholesterol: the Veterans Affairs HDL Intervention Trial
37. Shepherd J, Cobbe SM, Ford I, et al. Prevention of coronary heart dis-
(VA-HIT). Diabetes Care 2003;26:1513–7.
ease with pravastatin in men with hypercholesterolemia. West of
16. Prueksaritanont T, Tang C, Qiu Y, Mu L, Subramanian R, Lin JH. Effects
Scotland Coronary Prevention Study Group. N Engl J Med 1995;333:
of fibrates on metabolism of statins in human hepatocytes. Drug Metab
1301–7.
Dispos 2002;30:1280 –7.
38. Downs JR, Clearfield M, Weis S, et al. Primary prevention of acute
17. Pan WJ, Gustavson LE, Achari R, et al. Lack of a clinically significant
coronary events with lovastatin in men and women with average choles-
pharmacokinetic interaction between fenofibrate and pravastatin in
healthy volunteers. J Clin Pharmacol 2000;40:316 –23. terol levels: results of AFCAPS/TexCAPS. Air Force/Texas Coronary
18. Clofibrate and niacin in coronary heart disease. JAMA 1975;231:360 – 81. Atherosclerosis Prevention Study. JAMA 1998;279:1615–22.
19. Canner PL, Berge KG, Wenger NK, et al. Fifteen year mortality in 39. Jenkins DJ, Kendall CW, Marchie A, et al. Effects of a dietary portfolio
Coronary Drug Project patients: long-term benefit with niacin. J Am Coll of cholesterol-lowering foods vs lovastatin on serum lipids and C-reactive
Cardiol 1986;8:1245–55. protein. JAMA 2003;290:502–10.
20. Brown G, Albers JJ, Fisher LD, et al. Regression of coronary artery 40. Rubins HB, Robins SJ, Collins D, et al. Gemfibrozil for the secondary
disease as a result of intensive lipid-lowering therapy in men prevention of coronary heart disease in men with low levels of high-
with high levels of apolipoprotein B. N Engl J Med 1990;323: density lipoprotein cholesterol. Veterans Affairs High-Density
1289 –98. Lipoprotein Cholesterol Intervention Trial Study Group. N Engl J Med
21. Brown BG, Zhao XQ, Chait A, et al. Simvastatin and niacin, antioxidant 1999;341:410 – 8.
vitamins, or the combination for the prevention of coronary disease. 41. Schwartz GG, Olsson AG, Ezekowitz MD, et al., Myocardial Ischemia
N Engl J Med 2001;345:1583–92. Reduction with Aggressive Cholesterol Lowering (MIRACL) Study
22. Bays HE, Dujovne CA, McGovern ME, et al., ADvicor Versus Other Investigators. Effects of atorvastatin on early recurrent ischemic events in
Cholesterol-Modulating Agents Trial Evaluation. Comparison of once- acute coronary syndromes: the MIRACL study: a randomized controlled
daily, niacin extended-release/lovastatin with standard doses of atorva- trial. JAMA 2001;285:1711– 8.
732 Grundy et al. JACC Vol. 44, No. 3, 2004
Recent Clinical Trials and NCEP ATP III August 4, 2004:720–32

42. Goldberg RB, Mellies MJ, Sacks FM, et al. Cardiovascular events and 44. Pearson TA, Mensah GA, Alexander RW, et al., Centers for Disease
their reduction with pravastatin in diabetic and glucose-intolerant myo- Control and Prevention, American Heart Association. Markers of inflam-
cardial infarction survivors with average cholesterol levels: subgroup mation and cardiovascular disease: application to clinical and public
analyses in the cholesterol and recurrent events (CARE) trial. The Care health practice: a statement for healthcare professionals from the Centers
Investigators. Circulation 1998;98:2513–9. for Disease Control and Prevention and the American Heart Association.
43. Haffner SM, Alexander CM, Cook TJ, et al. Reduced coronary events Circulation 2003;107:499 –511.
in simvastatin-treated patients with coronary heart disease and 45. Jones P, Kafonek S, Laurora I, Hunninghake D. Comparative dose
diabetes or impaired fasting glucose levels: subgroup analyses in the efficacy study of atorvastatin versus simvastatin, pravastatin, lovastatin,
Scandinavian Simvastatin Survival Study. Arch Intern Med 1999;159: and fluvastatin in patients with hypercholesterolemia (the CURVES
2661–7. study). Am J Cardiol 1998;81:582–7.