Aspirin for the prevention of cardiovascular events in

patients without clinical cardiovascular disease:

A meta-analysis of randomized trials
Jeffrey S. Berger, MD, MS, FACC, a,b,c Anuradha Lala, MD, a Mori J. Krantz, MD, d,e
Gizelle S. Baker, PhD, f and William R. Hiatt, MD d,e New York, NY; and Denver, Colorado

Background The benefit of aspirin to prevent cardiovascular events in subjects without clinical cardiovascular disease
relative to the increased risk of bleeding is uncertain.
Methods A meta-analysis of randomized trials of aspirin versus placebo/control to assess the effect of aspirin on major
cardiovascular events (MCEs) (nonfatal myocardial infarction, nonfatal stroke, or cardiovascular death), individual
components of the MCE, stroke subtype, all-cause mortality, and major bleeding. Nine trials involving 102,621 patients
were included: 52,145 allocated to aspirin and 50,476 to placebo/control.
Results Over a mean follow-up of 6.9 years, aspirin was associated with a reduction in MCE (risk ratio [RR] 0.90, 95% CI
0.85-0.96, P b .001). There was no significant reduction for myocardial infarction, stroke, ischemic stroke, or all-cause
mortality. Aspirin was associated with hemorrhagic stroke (RR 1.35, 95% CI 1.01-1.81, P = .04) and major bleeding (RR
1.62, 95% CI 1.31-2.00, P b .001). In meta-regression, the benefits and bleeding risks of aspirin were independent of
baseline cardiovascular risk, background therapy, age, sex, and aspirin dose. The number needed to treat to prevent 1 MCE
over a mean follow-up of 6.9 years was 253 (95% CI 163-568), which was offset by the number needed to harm to cause
1 major bleed of 261 (95% CI 182-476).
Conclusions The current totality of evidence provides only modest support for a benefit of aspirin in patients without
clinical cardiovascular disease, which is offset by its risk. For every 1,000 subjects treated with aspirin over a 5-year period,
aspirin would prevent 2.9 MCE and cause 2.8 major bleeds. (Am Heart J 2011;162:115-124.e2.)

Aspirin is effective in decreasing cardiovascular mor-
bidity and mortality in patients with clinical evidence of
cardiovascular disease (CVD).1,2 Specifically, benefits of
aspirin are well defined for secondary prevention of acute
myocardial infarction (MI) and stroke, forming the basis
for current clinical practice guidelines.3-6 These respon-
From the aDepartment of Medicine, Division of Cardiology, New York University School of
Medicine, New York, NY, bDivision of Hematology, New York University School of
Medicine, New York, NY, cDivision of Vascular Surgery, Department of Surgery, New York
University School of Medicine, New York, NY, dDivision of Cardiology, Department of
Medicine, University of Colorado Denver School of Medicine, Denver, Colorado, eSection
of Vascular Medicine, Colorado Prevention Center, Denver, Colorado, and fDepartment of
Biostatistics and Informatics, Colorado School of Public Health, Denver, Colorado.
Roger S. Blumenthal, MD served as guest editor for this article.
Dr Berger was partially funded by an American Heart Association Fellow to Faculty Award
(0775074N). Dr Krantz was partially funded by grant U01 HL079160 from the National
Heart, Lung, and Blood Institute.
Submitted March 3, 2011; accepted April 6, 2011.
Reprint requests: Jeffrey S. Berger, MD, MS, FACC, The Leon H Charney Division of
Cardiology, New York University School of Medicine, New York, NY 10016.
E-mail: Jeffrey.berger@nyumc.org
0002-8703/$ - see front matter
2011, Mosby, Inc. All rights reserved.
doi:10.1016/j.ahj.2011.04.006
sive populations include patients who have experienced
plaque rupture or vessel occlusion sufficient to induce a
symptomatic state (eg, stable angina7 or transient
ischemic attack),8,9 acute coronary syndrome,10-12 or
ischemic stroke.13 The data demonstrating benefit of
aspirin in patients without clinical CVD are less cer-
tain.14,15 This includes populations at risk based on age,
risk factors such as diabetes or hypertension, or evidence
of subclinical atherosclerosis.16 Clinical decision making
is further complicated by several meta-analyses of the
older trials that found a significant decrease in the
composite of major cardiovascular events (MCEs) by
approximately 12%,17,18 but newer trials of aspirin in
patients at high risk but without clinical CVD have failed
to achieve their primary end points.19-21 Thus, the
benefit-to-risk relationship needs further definition in
light of the newer evidence.
The most recent meta-analysis reviewed 6 primary
prevention trials and concluded that current evidence did
not clearly support a net benefit of aspirin relative to the
excess bleeding risk in primary prevention.17 This
analysis did not identify any potential subgroups as
response modifiers including age, gender, diabetes or
 American Heart Journal
116 Berger et al July 2011

Table I. Design of trials included in the meta-analysis

PHS BDT TPT HOT PPP WHS POPADAD JPAD AAA

Year of 1989 1988 1998 1998 2001 2005 2008 2008 2010
publication
No. of 22071 5139 5085 18790 4495 39876 1276 2539 3350
participants
Patient Healthy Healthy Men at Men and Men and Healthy Men and Men and Men and
population male male high risk women with women N1 females women with women with women
doctors doctors for CVD hypertension cardiovascular diabetes and diabetes in general
risk factor ABI 0.99 population
with ABI
0.95
Country USA UK UK UK, USA, Italy USA UK Japan UK
Asia
Mean year 5 5.6 6.7 3.8 3.7 10 6.7 4.4 8.2
follow-up
Aspirin dose 325 mg 500 mg 75 mg daily 75 mg daily 100 mg 100 mg 100 mg 81 mg or 100 mg daily
every daily or (controlled daily every daily 100 mg daily (enteric coated)
other day 300 mg release) other day
Primary CV death CV death, All ischemic CV death, CV death, CV death, CV death, All ischemic CV death,
end point nonfatal MI, heart disease nonfatal MI, stroke nonfatal MI, nonfatal MI, heart disease, MI, stroke,
stroke, TIA (coronary death MI, stroke stroke stroke, stroke, and revascularization
and fatal and critical limb peripheral
nonfatal MI) ischemia artery disease

CV, Cardiovascular; TIA, transient ischemic attack.

Median year follow-up.
If requested.

other risk factors, or predicted 5-year risk of CVD. Outcome measures

However, the publication of 3 new trials in at-risk
populations allows for further evaluation of these
potential risk modifiers. Thus, the current meta-analysis
includes new data and tests the null hypothesis on the
available data that there is no net benefit relative to risk
for aspirin when used in patients without clinical CVD.
Methods
Search strategy
From 1966 to 2005, a computerized search was performed
that identified 6 published randomized trials of aspirin in
patients without clinical CVD. In the previous meta-analysis of
these trials, aspirin therapy significantly reduced the risk of an
MCE (nonfatal MI, nonfatal stroke, or cardiovascular death) by
12% in women and 14% in men.18 From 2005 to the present, a
subsequent review of the literature (MEDLINE, the Cochrane
Central Register of Controlled Trials, and EMBASE) identified 3
additional primary prevention studies of aspirin for a total of 9
trials for analysis.19-21
Study selection
Primary prevention trials that involved a randomized com-
parison of aspirin versus placebo or control in patients without
clinical CVD (eg, established or symptomatic) were included in
this analysis. The criteria for inclusion of trials were as follows:
(1) aspirin alone was used for the primary prevention of CVD;
(2) comparisons of outcomes were made between aspirin and
placebo or open control groups; and (3) data were available on
MI, stroke, and cardiovascular deaths. Our search included only
those studies published in English.
The primary outcome was the risk ratio (RR) of aspirin therapy
compared with placebo or control on the composite MCE end
point, which includes nonfatal MI, nonfatal stroke, or cardiovas-
cular death. Secondary outcomes included all MI, all stroke, all-
cause mortality, and cardiovascular mortality. The primary safety
outcome was the occurrence of major bleeding as defined by
each study. Because the definition of major bleeding differed by
trial, gastrointestinal hemorrhage and cerebral hemorrhage were
reported separately. In addition, stroke subtypes (ischemic or
hemorrhagic) were examined from data available in the 8 studies
that reported strokes by subtype. The adjudication process to
subtype stroke was not provided for all the trials. The HOT trial22
did not include data by stroke subtypes.
Statistical analysis
The composite end point was analyzed using the data as they
were reported in the included publications. A random-effects
model was the primary analysis because of the differences
across the studies in patient characteristics (including underly-
ing disease population) and aspirin dosing. To determine the
appropriateness of pooled RR estimates, heterogeneity among
the trials was examined using the Q statistic for each of the end
points analyzed. It was determined that pooled estimates from
the random-effects model of the RR were reasonable because
the Q statistic did not indicate statistically significant heteroge-
neity and the I2 statistic was b37% for all end points other than
fatal and nonfatal MI. For the MI end point, there was moderate
heterogeneity (63% of variation because of heterogeneity) that
was statistically significant.
A sensitivity analysis of the primary analysis was performed
to examine the robustness of the results. This was done by
examining the impact of systematically removing 1 study from
American Heart Journal
Volume 162, Number 1
Berger et al 117

Table II. Patient characteristics included in the meta-analysis

PHS BDT TPT HOT PPP WHS POPADAD JPAD AAA

Age (mean) 53 (9) 50-78 58 62 64 (8) 55 (7) 60 65 62

Female (%) 0 0 0 47 58 100 56 45 72
Mean SBP (mm Hg) 126 136 139 170 145 b120/75, 32%; 145 135 148
120-9/75-84, 32%;
130-9/85-89, 19%;
N140/90, 16%
Diabetes (%) 2.4 2 NA 8 17 2.6 100 100 3
Total cholesterol (mg/dL) 212 NA 247 236 236 NA 213 201 240
Smokers (%) 11 13 41 16 15 13 31 21 65
BMI (kg/m2) 24.9 (3) NA 27.4 28.4 27.6(5) 26 (5) 24 (4) 29 NA
Annual risk of 0.67 1.54 1.53 1.19 0.76 0.26 2.53 0.82 0.99
cardiovascular events (%)

SBP, Systolic blood pressure; BMI, body mass index.

the analysis and recalculating the results. In addition, specific
subsets of the studies were analyzed to evaluate the consistency
of the effects of aspirin when studies that were restricted to
persons with diabetes were excluded,19,21 studies enrolling
patients with subclinical atherosclerosis (by a low ankle-
brachial index [ABI]) were excluded,19,20 and studies that
included extended or controlled release aspirin were re-
moved.20,23,24 In addition, the data were analyzed using a
fixed-effect model, and the results were compared with the
random-effects model.
Linear meta-regression analyses on the log-transformed RR with
the study as the unit of analysis were performed to evaluate
potential effect modifiers (year of study publication, baseline
cardiovascular risk as assessed by incidence of events on placebo,
mean age of the trial participant, sex, and dose of aspirin).
Studies were weighed by the precision of the effect estimate.
The potential publication bias was examined by constructing
a funnel plot in which sample size was plotted against log RRs
for the primary end point available from all studies.
Statistical analyses were performed using Review Manager
5.0.23 (The Nordic Cochrane Centre, The Cochrane Collabora-
tion, Copenhagen, Denmark, 2008). P b .05 was judged as
statistically significant.
Results
Nine prospective randomized trials involving 102,621
participants were identified for inclusion. A total of
710,053 person-years of exposure were recorded:
359,709 in the aspirin group and 350,344 in the placebo
or control group. All trials included patients without
clinical CVD, which was defined as the absence of a
cardiovascular event, or clinical symptoms of CVD
including angina or transient ischemic attack. Among
the 3 new trials,19-21 2 included only diabetic pa-
tients,19,21 and 2 required a low ABI measurement as a
marker of subclinical atherosclerosis for inclusion.19,20
Of the 9 trials, 3 included only men,23-25 and 1 included
only women.26
Details of the included studies appear in Table I, and
characteristics of participants are summarized in Table II.
Dosage of aspirin ranged from 100 mg every other day to
500 mg daily. Two trials used enteric-coated aspirin,20,24
and 1 trial used a longer preparation tablet.23 Follow-up
ranged from 4.4 to 10.1 years and was N95% complete in
all trials. The weighted mean follow-up time was
6.9 years.
Major cardiovascular events
A total of 2,029 MCEs occurred among 52,145 (3.86%)
patients allocated to aspirin compared with 2,099 MCEs
among 50,476 (4.16%) patients assigned to placebo or
control (Figure 1). Pooled results using a random-effects
model demonstrated a significant 10% reduction in the
risk of MCE (RR 0.90, 95% CI 0.85-0.96, P b .001). In
aggregate, the absolute risk reduction was 0.39% (95% CI
0.18%-0.61%) over a mean follow-up of 6.9 years, which
corresponds to a number needed to treat of 253 (95% CI
163-568) to prevent a single MCE.
Myocardial infarction
A total of 993 fatal and nonfatal MI events occurred
among 52,145 (1.90%) patients allocated to aspirin
compared with 1,075 MI events among 50,476 (2.13%)
patients assigned to placebo or control (Figure 2). Pooled
results demonstrated a 14% risk reduction in both fatal
and nonfatal MI, which failed to reach statistical
significance (RR 0.86, 95% CI 0.74-1.00, P = .06). For
the end point of nonfatal MI, there was a 16% reduction in
events with aspirin versus placebo or control; however,
the results did not reach statistical significance (RR 0.84,
95% CI 0.70-1.01, P = .06).
Stroke
A total of 741 fatal and nonfatal strokes occurred among
52,145 (1.42%) patients allocated to aspirin compared
with 746 strokes among 50,476 (1.48%) patients assigned
to placebo or control (Figure 3). Pooled results demon-
strated a nonsignificant 6% reduction in the risk of stroke
(RR 0.94, 95% CI 0.84-1.06, P = .31). Eight of the studies

118 Berger et al
Figure 1
Effect of aspirin on the prevention of MCE. Major cardiovascular events defined as the composite of nonfatal MI, nonfatal stroke, or
cardiovascular death.
Figure 2
Effect of aspirin on the prevention of MI (fatal and nonfatal).
provided data on type of stroke. For the end point of
nonfatal stroke, there was a nonsignificant 8% reduction
in events with aspirin versus placebo or control (RR 0.92,
95% CI 0.80-1.06, P = .26).
To further understand the benefit versus risk, the
relationship of aspirin on ischemic and hemorrhagic
stroke was explored independently (Figure 3). For the
end point of ischemic stroke, there was a nonsignificant
13% reduction in events with aspirin versus placebo or
control (RR 0.87, 95% CI 0.73-1.02, P = .09). For the end
point of hemorrhagic stroke, there was a significant 35%
increase in events with aspirin versus placebo or control
(RR 1.35, 95% CI 1.01-1.81, P = .04). The absolute risk
increase was 0.06% (95% CI 0.003-0.126) over a mean
follow-up of 6.9 years, corresponding to a number
needed to harm of 1,560 (95% CI 794-33,333) to cause
1 hemorrhagic stroke.
Death
A total of 1,962 deaths occurred among 52,145
(3.76%) patients allocated to aspirin compared with
American Heart Journal
July 2011
1,933 among 50,476 (3.83%) patients assigned to
placebo or control (Figure 4). Pooled results demon-
strated a nonsignificant reduction of 6% in the risk of
death (RR 0.94, 95% CI 0.89-1.00, P = .07). For the end
point of cardiovascular death, there was no reduction in
events with aspirin versus placebo or control (RR 0.99,
95% CI 0.85-1.14, P = .86).
Major bleeding
A total of 458 major bleeding events occurred among
52,145 (0.88%) patients allocated to aspirin compared
with 278 major bleeding events among 50,421 (0.55%)
patients assigned to placebo or control (Figure 5).
Pooled results demonstrated a significant 62% increase
in the risk of major bleeding (RR 1.62, 95% CI 1.31-2.00,
P b .001). In aggregate, the absolute risk was 0.38%
(95% CI 0.21%-0.55%) over a mean follow-up of 6.9
years, which corresponds to a number needed to harm
of 261 (95% CI 182-476) to cause a single major bleeding
event over 6.9 years. For the end point of gastrointes-
tinal hemorrhage, there was a significant 29% increase
American Heart Journal
Volume 162, Number 1
Figure 3
Effect of aspirin on the prevention of stroke (fatal and nonfatal), ischemic stroke, and hemorrhagic stroke.
in events with aspirin versus placebo or control (RR
1.29, 95% CI 1.24-1.47, P b .001).
Meta-regression
Linear meta-regression models were used to assess
the impact of study-level covariates on the risk of MCE
and major bleeding. There was no relationship between
any of the covariates examined (year of study
publication as a surrogate for any changes in back-
ground cardiovascular preventive therapies, baseline
cardiovascular risk as assessed by incidence of events
on placebo, mean age of the trial participant, sex of the
trial participant, and dose of aspirin) and the effect of
Berger et al 119
aspirin on MCE or major bleeding (see Supplemental
Figures 6 and 7 in the online Appendix).
Aspirin dose and formulation
Although aspirin dose ranged from 100 mg every
other day to 500 mg/d, 7 of the 9 trials used a dose
between 75 and 162.5 mg/d. There did not appear to be
any relationship between aspirin dose and the effect of
aspirin on MCE or major bleeding (Supplemental Figures
6E and 7E in the online Appendix). Three trials used a
different aspirin formulation than regular aspirin. The
BDT and AAA trials used an enteric-coated aspirin
formulation, whereas the TPT trial used a controlled

120 Berger et al
Figure 4
Effect of aspirin on the prevention of all-cause death and cardiovascular death.
release formulation. When analyses were repeated after
excluding these trials, the association between aspirin
and the primary outcome of MCE yielded an effect size
(RR 0.89, 95% CI 0.83-0.95) similar in magnitude and
direction to the overall results.
Sensitivity analysis
There was no difference in results between the fixed-
effect (RR 0.90) or the random-effect model (RR 0.90) for
the primary outcome of MCE. The RR varied between
0.891 and 0.893 when each study was systematically
removed from the model. The POPADAD19 and JPAD21
trials exclusively enrolled patients with diabetes, and
both demonstrated no significant effect on the composite
of cardiovascular events. Because aspirin may not be
beneficial in patients with diabetes,27 the primary analysis
was repeated after excluding these 2 trials. The
association between aspirin and the primary outcome
of MCE yielded an effect size (1,868 events among 50,245
patients taking aspirin [3.7%] vs 1,924 events among
48,561 control patients [4.0%]; RR 0.90, 95% CI 0.84-0.96)
identical to the overall results.
The AAA20 and POPADAD19 trials exclusively enrolled
patients with subclinical atherosclerosis determined by a
American Heart Journal
July 2011
low ABI. Because aspirin may not be beneficial in patients
with PAD,28 the primary analysis was repeated after
excluding these 2 trials. A similar effect of aspirin was
observed (RR 0.89, 95% CI 0.83-0.95) relative to the
overall result.
Discussion
The current meta-analysis included 9 prevention trials
in N100,000 subjects without clinical evidence of CVD
treated with aspirin versus placebo or control for the
prevention of ischemic cardiovascular events. The
pooled results found a statistically significant 10% relative
reduction in the primary end point of MCE and a
statistically significant 62% relative increase in major
bleeding events. The absolute benefit versus risk
demonstrated that, in 1,000 patients treated for 5 years,
there were approximately 3 ischemic events avoided
that was associated with an excess of 3 major bleeds
resulting in no net benefit to risk for aspirin in this
population. These results were consistent across a
number of subgroups suggesting that a more targeted
approach to maximize benefit and minimize risk could
not be identified.
The year of publication of the trials ranged from 1988 to
2010 and spanned over a decade during which there
American Heart Journal
Volume 162, Number 1
Figure 5
Effect of aspirin on major bleeding.
were major improvements in the medical prevention and
treatment of CVD,29,30 yet publication year did not have
either a positive or negative impact on the overall
estimate of the benefit or risk of aspirin (Supplemental
Figures 6A and 7A in the online Appendix). The risk of
CVD events in the placebo/control groups was also quite
broad, ranging from 0.25% per year in the Women's
Health Study to 2.7% in the POPADAD trial (expressed
over a decade per the Framingham coronary heart
disease risk score this baseline risk ranges from 2.5% to
27%). Yet, using this baseline risk assessment or
components of that risk including populations enriched
with diabetes did not impact the estimate of benefit or
risk of aspirin (Supplemental Figures 6B and 7B in the
online Appendix). Cardiovascular risk estimates can also
be significantly influenced by measures of subclinical
atherosclerosis such as the ABI or carotid intima-media
thickness.31-33 However, in the most recent AAA trial20
where a low ABI was used as an inclusion criterion or in
the POPADAD trial19 where a reduced ABI and diabetes
were inclusion criteria, the trials were not just statisti-
cally negative, but the point estimates of benefit in both
were near 1.00. Finally, although the dose of aspirin
ranged from 100 mg every other day (50 mg/d)26 to 500
mg/d,24 the dose of aspirin did not impact the results
(Supplemental Figures 6E and 7E in the online Appendix).
Taken together, aspirin retains a modest yet statistically
significant benefit on decreasing nonfatal cardiovascu-
lar events. These benefits are offset by a consistent
signal of major bleeding risk including hemorrhagic
stroke. These new results need to be considered by
clinicians making difficult decisions about the value of
primary preventive pharmacotherapies (relative to
drugs with unequivocal benefit such as statins and
blood pressure medications).
The goal of the current study was to perform a
comprehensive meta-analysis of all randomized trials
with aspirin as preventive therapy in patients without
established or symptomatic CVD. Previous meta-ana-
Berger et al 121
lyses17,18,34 have been restricted to older trials with
lower risk populations and did not include all contempo-
rary randomized trials. Twenty years ago, 2 landmark
trials24,25 evaluated the effect of aspirin in primary
prevention. Neither the BDT24 nor the PHS25 demon-
strated a benefit for aspirin in the reduction of their
primary end point (Table III). However, the PHS25 was
able to demonstrate a decrease in the secondary end
point of fatal and nonfatal MI by 44%, leading to the
widespread recommendations and use of aspirin in
primary prevention. Since then, there have been 4
randomized trials22,23,26,35 of aspirin in primary preven-
tion and 3 more recent randomized trials19-21 in patients
with diabetes and/or subclinical atherosclerosis (defined
by reduced ABI). None of these 7 trials demonstrated a
beneficial effect of aspirin on their prespecified primary
end point (Table III).16 If one were to evaluate the
traditional composite end point of nonfatal MI, nonfatal
stroke, or cardiovascular death, only 1 of the 9 trials was
associated with a statistically significant reduction. Thus,
the benefit observed in our meta-analysis with aspirin is
only seen after combining all the data in aggregate.
Bleeding risk is consistently increased with aspirin
therapy across the trials. Despite different definitions of
major bleeding among the trials, there was a strong signal
of increased risk. Of the 9 trials, 4 found a statistically
significant increase in major bleeding, and in pooled
analysis, the risk was increased by 62%. Hemorrhagic
stroke is a less common but clinically important end point
because of its significant morbidity. Incidence of
hemorrhagic stroke increased by approximately 0.1%
(1 per 1,000 patients treated); however, in the pooled
analysis, aspirin was associated with a significant 36%
increase in the risk of hemorrhagic stroke.
Because aspirin was associated with a significant
reduction in cardiovascular events but a proportional
increase in bleeding complications, it is essential to
understand the absolute benefit versus risk. Three MCEs
were prevented for every 1,000 patients treated with
 American Heart Journal
122 Berger et al July 2011

Table III. Primary efficacy end point for each trial fibrates,39,43 phosphodiesterase inhibitors such as cilos-
Primary efficacy
tazol,44 and selective serotonin reuptake inhibitors45
Study, year end point HR (95% CI) P have all been shown to reduce platelet activity.
Unfortunately, few of the aspirin trials reported
PHS, 1989 Cardiovascular RR 0.96 .87 concomitant medications and what, if any, modifying
mortality (0.60-1.54) effect these medications may have on the effect of
BDT, 1988 CV death, nonfatal MI NS
aspirin. Although some have suggested an attenuated
and stroke or TIA
TPT, 1998 All ischemic heart disease 20% (1-35) .04 effect of aspirin in certain high-risk groups, such as
(coronary death and reduction persons with diabetes 27 and peripheral artery disease,28
fatal and nonfatal MI) there was no difference in the results when trials that
HOT, 1998 CV death, nonfatal MI, stroke RR 091 .17 exclusively enrolled those populations were excluded.
(079-104)
Finally, dose and aspirin formulation differed across the
PPP, 2001 CV death, MI, stroke 071 (048-104) NS
WHS, 2005 CV death, nonfatal MI, stroke 0.91 (0.80-1.03) .13 trials, which may have contributed to the lower
POPAD, CV death, nonfatal MI or 0.98 (0.76-1.26) .86 proportional benefit. However, there was no difference
2008 stroke, or amputation for in the effect of aspirin when adjusting for aspirin dose
critical limb ischemia or after excluding trials that used a longer acting or
JPAD, 2008 CV death, nonfatal MI, 0.80 (0.58-1.10) .16
stroke, UA, PVD, new angina
enteric-coated formulation.
AAA, 2010 CV death, MI, stroke, 1.03 (0.84-1.27) NS
revascularization Limitations
NS, Not significant. There are several limitations to this study. Some
After including silent MI, the reduction was no longer significant (P = .07). imprecision exists in the frequency of events because
After excluding silent MI, the reduction became statistically significant (P = .03).
the protocols for data collection and definitions of
efficacy and safety events varied among the studies.
However, all of the trials allowed for the assessment of a
aspirin over a period of 5 years, at the cost of inducing 3 common MCE outcome that served as a primary end
major bleeding events. Intracranial hemorrhage was point for analysis. For the safety end point, definitions of
statistically significantly higher in the aspirin group, but major bleeding varied across the trials, and these
the number needed to harm was 1,560 (b1 in 1,000 variations made it difficult to determine accurate
patients treated). measure of bleeding and approximate risk. Therefore,
There are several potential explanations for the lower- both gastrointestinal hemorrhage and hemorrhagic
than-expected proportional benefit observed with aspirin stroke were used as complementary end points to help
in this trial data set. Trials included in this analysis evaluate risk.
spanned several decades, and during this period, there The current meta-analysis was not performed on
has been considerable improvement of outcomes with individual patient data because complete individual
the adoption of statins and angiotensin-converting data were not available, and therefore, comprehensive
enzyme inhibitors.29,36 Although no significant heteroge- assessment within different higher-risk groups was not
neity was observed between proportional reduction in able to be performed. Nonetheless, the analyses were
MCE and time of publication, 4 of 5 trials published able to adjust for the incidence of adverse events in the
before 2002 had a proportional reduction of N10%, placebo group (a surrogate for baseline risk) in each
whereas only 1 of 4 trials published after 2003 had such a individual trial, which had no effect on the overall
benefit. Moreover, the larger proportional benefit (N15%) results. A proportion of patients included in this meta-
observed in trials of aspirin in patients with clinical CVD analysis had documented subclinical atherosclerosis, as
occurred in trials N15 years ago. Consistent with this detected by a lower ABI. Although some would consider
hypothesis, the CHARISMA trial (published in 2006) this a limitation because of the higher-risk cohort
tested clopidogrel plus aspirin versus aspirin alone in included, no change in the effect of aspirin was observed
patients with and without clinical CVD and demonstrated after exclusion of these patients, and the results were
a nonsignificant increased risk of MCE with combination nearly identical.
antiplatelet therapy in the primary prevention subgroup
of the trial.37 Thus, if aspirin alone is of uncertain benefit
in these subjects, dual antiplatelet therapy may be even Conclusions
more suspect. This meta-analysis of N100,000 randomized patients
An additional explanation on the lack of a net (N700,000 person-year follow-up) comparing aspirin
benefit over risk may be that concomitant therapies versus placebo or control demonstrated that aspirin
have minor antiplatelet affects that may attenuate any decreased MCE by approximately 10% among patients
potential benefit of aspirin. In fact, statins,38,39 fish without clinical CVD. Major bleeding, however, occurred
oil,40 angiotensin-converting enzyme inhibitors,41,42 more frequently with aspirin therapy. The decision to use
American Heart Journal
Volume 162, Number 1
aspirin for the prevention of a first MI or stroke remains a
complex issue. Weighing the overall benefit and risk
requires careful consideration by the physician and
patient before initiating aspirin for preventive therapy
in patients without clinical CVD.
Acknowledgements
We thank Dr Sanjay Kaul for his insightful comments
and editorial assistance.
Disclosures
Dr Jeffrey Berger reports receiving research support
from Astra Zaneca and has received honoraria for advisory
board participation from Astra Zaneca. Dr William Hiatt
reports receiving research support from Astra Zaneca.
Conflicts of interest: none. Drs Berger and Hiatt had
full access to all the data in the study and take
responsibility for the integrity of the data and the
accuracy of the data analysis.
References
1. Collaborative meta-analysis of randomised trials of antiplatelet
therapy for prevention of death, myocardial infarction, and stroke in
high risk patients. BMJ (Clin Res ed) 2002;324:71-86.
2. Berger JS, Brown DL, Becker RC. Low-dose aspirin in patients with
stable cardiovascular disease: a meta-analysis. Am J Med 2008;121:
43-9.
3. Patrono C, Baigent C, Hirsh J, et al. Antiplatelet drugs: American
College of Chest Physicians evidence-based clinical practice guide-
lines (8th Edition). Chest 2008;133:199S-233S.
4. Smith Jr SC, Allen J, Blair SN, et al. AHA/ACC guidelines for
secondary prevention for patients with coronary and other athero-
sclerotic vascular disease: 2006 update: endorsed by the National
Heart, Lung, and Blood Institute. Circulation 2006;113:2363-72.
5. Antman EM, Hand M, Armstrong PW, et al. 2007 focused update of
the ACC/AHA 2004 guidelines for the management of patients with
ST-elevation myocardial infarction: a report of the American College
of Cardiology/American Heart Association Task Force on Practice
Guidelines: developed in collaboration with the Canadian Cardio-
vascular Society endorsed by the American Academy of Family
Physicians: 2007 writing group to review new evidence and update
the ACC/AHA 2004 guidelines for the management of patients with
ST-elevation myocardial infarction, writing on behalf of the 2004
writing committee. Circulation 2008;117:296-329.
6. Anderson JL, Adams CD, Antman EM, et al. ACC/AHA 2007
guidelines for the management of patients with unstable angina/non
ST-elevation myocardial infarction: a report of the American College
of Cardiology/American Heart Association Task Force on Practice
Guidelines (writing committee to revise the 2002 guidelines for the
management of patients with unstable angina/non ST-elevation
myocardial infarction): developed in collaboration with the American
College of Emergency Physicians, the Society for Cardiovascular
Angiography and Interventions, and the Society of Thoracic
Surgeons: endorsed by the American Association of Cardiovascular
and Pulmonary Rehabilitation and the Society for Academic
Emergency Medicine. Circulation 2007;116:e148-304.
Berger et al 123
7. Juul-Moller S, Edvardsson N, Jahnmatz B, et al. Double-blind trial of
aspirin in primary prevention of myocardial infarction in patients with
stable chronic angina pectoris. The Swedish Angina Pectoris Aspirin
Trial (SAPAT) group. Lancet 1992;340:1421-5.
8. The SALT Collaborative Group. Swedish Aspirin Low-Dose Trial
(SALT) of 75 mg aspirin as secondary prophylaxis after cerebrovas-
cular ischaemic events. Lancet 1991;338:1345-9.
9. Farrell B, Godwin J, Richards S, et al. The United Kingdom transient
ischaemic attack (UK-TIA) aspirin trial: final results. J Neurol
Neurosurg Psychiatry 1991;54:1044-54.
10. ISIS-2 (Second International Study of Infarct Survival) Collaborative
Group. Randomised trial of intravenous streptokinase, oral aspirin,
both, or neither among 17,187 cases of suspected acute myocardial
infarction: ISIS-2. Lancet 1988;2:349-60.
11. Lewis Jr HD, Davis JW, Archibald DG, et al. Protective effects of
aspirin against acute myocardial infarction and death in men with
unstable angina. Results of a Veterans Administration Cooperative
Study. N Engl J Med 1983;309:396-403.
12. Wallentin LC. Aspirin (75 mg/day) after an episode of unstable
coronary artery disease: long-term effects on the risk for myocardial
infarction, occurrence of severe angina and the need for revascu-
larization. Research Group on Instability in Coronary Artery Disease
in Southeast Sweden. J Am Coll Cardiol 1991;18:1587-93.
13. Chen ZM, Sandercock P, Pan HC, et al. Indications for early aspirin
use in acute ischemic stroke : a combined analysis of 40 000
randomized patients from the Chinese acute stroke trial and the
international stroke trial. On behalf of the CAST and IST collaborative
groups. Stroke 2000;31:1240-9.
14. Krantz MJ, Berger JS, Hiatt WR. An aspirin a day: are we barking up
the wrong willow tree? Pharmacotherapy 2010;30:115-8.
15. Berger JS. Aspirin as preventive therapy in patients with asymptom-
atic vascular disease. JAMA 2010;303:880-2.
16. Cardiovascular and Renal Drugs Advisory Committee. Available at:
http://www.fda.gov/ohrms/DOCKETS/ac/03/ques-
tions/4012Q1_Final.pdf. Last accessed on March 13, 2010.
17. Baigent C, Blackwell L, Collins R, et al. Aspirin in the primary and
secondary prevention of vascular disease: collaborative meta-
analysis of individual participant data from randomised trials. Lancet
2009;373:1849-60.
18. Berger JS, Roncaglioni MC, Avanzini F, et al. Aspirin for the primary
prevention of cardiovascular events in women and men: a sex-
specific meta-analysis of randomized controlled trials. JAMA 2006;
295:306-13.
19. Belch J, MacCuish A, Campbell I, et al. The prevention of progression
of arterial disease and diabetes (POPADAD) trial: factorial rando-
mised placebo controlled trial of aspirin and antioxidants in patients
with diabetes and asymptomatic peripheral arterial disease. BMJ
(Clin Res ed) 2008;337:a1840.
20. Fowkes FG, Price JF, Stewart MCW, et al. Aspirin for prevention of
cardiovascular events in a general population screened for a low
ankle brachial index. JAMA 2010;303:841-8.
21. Ogawa H, Nakayama M, Morimoto T, et al. Low-dose aspirin for
primary prevention of atherosclerotic events in patients with type 2
diabetes: a randomized controlled trial. JAMA 2008;300:2134-41.
22. Hansson L, Zanchetti A, Carruthers SG, et al. Effects of intensive
blood-pressure lowering and low-dose aspirin in patients with
hypertension: principal results of the Hypertension Optimal Treatment
(HOT) randomised trial. HOT Study Group. Lancet 1998;351:
1755-62.
23. The Medical Research Council's General Practice Research Frame-
work. Thrombosis prevention trial: randomised trial of low-intensity
oral anticoagulation with warfarin and low-dose aspirin in the

124 Berger et al
primary prevention of ischaemic heart disease in men at increased
risk. Lancet 1998;351:233-41.
24. Peto R, Gray R, Collins R, et al. Randomised trial of prophylactic daily
aspirin in British male doctors. Br Med J (Clin Res ed) 1988;296:
313-6.
25. Final report on the aspirin component of the ongoing Physicians'
Health Study. Steering Committee of the Physicians' Health Study
Research Group. N Engl J Med 1989;321:129-35.
26. Ridker PM, Cook NR, Lee IM, et al. A randomized trial of low-dose
aspirin in the primary prevention of cardiovascular disease in
women. N Engl J Med 2005;352:1293-304.
27. De Berardis G, Sacco M, Strippoli GF, et al. Aspirin for primary
prevention of cardiovascular events in people with diabetes: meta-
analysis of randomised controlled trials. BMJ (Clin Res ed 2009;339:
b4531.
28. Berger JS, Krantz MJ, Kittelson JM, et al. Aspirin for the prevention of
cardiovascular events in patients with peripheral artery disease: a
meta-analysis of randomized trials. JAMA 2009;301:1909-19.
29. Ford ES, Ajani UA, Croft JB, et al. Explaining the decrease in U.S.
deaths from coronary disease, 1980-2000. N Engl J Med 2007;356:
2388-98.
30. Hunink MG, Goldman L, Tosteson AN, et al. The recent decline in
mortality from coronary heart disease, 1980-1990. The effect of
secular trends in risk factors and treatment. JAMA 1997;277:
535-42.
31. Berger JS, Jordan CO, Lloyd-Jones D, et al. Screening for
cardiovascular risk in asymptomatic patients. J Am Coll Cardiol
2010;55:1169-77.
32. Fowkes FG, Murray GD, Butcher I, et al. Ankle brachial index
combined with Framingham Risk Score to predict cardiovascular
events and mortality: a meta-analysis. JAMA 2008;300:197-208.
33. O'Leary DH, Polak JF, Kronmal RA, et al. Carotid-artery intima and
media thickness as a risk factor for myocardial infarction and stroke
in older adults. Cardiovascular Health Study Collaborative Research
Group. N Engl J Med 1999;340:14-22.
34. Eidelman RS, Hebert PR, Weisman SM, et al. An update on aspirin in
the primary prevention of cardiovascular disease. Arch Intern Med
2003;163:2006-10.
35. de Gaetano G. Low-dose aspirin and vitamin E in people at
cardiovascular risk: a randomised trial in general practice. Collab-
American Heart Journal
July 2011
orative Group of the Primary Prevention Project. Lancet 2001;357:
89-95.
36. Cooper R, Cutler J, Desvigne-Nickens P, et al. Trends and
disparities in coronary heart disease, stroke, and other cardiovas-
cular diseases in the United States: findings of the national
conference on cardiovascular disease prevention. Circulation 2000;
102:3137-47.
37. Bhatt DL, Fox KA, Hacke W, et al. Clopidogrel and aspirin versus
aspirin alone for the prevention of atherothrombotic events. N Engl J
Med 2006;354:1706-17.
38. Haramaki N, Ikeda H, Takenaka K, et al. Fluvastatin alters platelet
aggregability in patients with hypercholesterolemia: possible im-
provement of intraplatelet redox imbalance via HMG-CoA reductase.
Arterioscler Thromb Vasc Biol 2007;27:1471-7.
39. Damas JK, Puccetti L, Aukrust P. Early anti-thrombotic and anti-
inflammatory actions of statins and fibratestime for adjuvant
therapy in acute coronary syndromes? Thromb Haemost 2005;94:
1-3.
40. Lee JH, O'Keefe JH, Lavie CJ, et al. Omega-3 fatty acids:
cardiovascular benefits, sources and sustainability. Nat Rev Cardiol
2009;6:753-8.
41. Okrucka A, Pechan J, Kratochvilova H. Effects of the angioten-
sin-converting enzyme (ACE) inhibitor perindopril on endothelial
and platelet functions in essential hypertension. Platelets 1998;9:
63-7.
42. Moser L, Callahan KS, Cheung AK, et al. ACE inhibitor effects on
platelet function in stages I-II hypertension. J Cardiovasc Pharmacol
1997;30:461-7.
43. Barter PJ, Rye KA. Cardioprotective properties of fibrates: which
fibrate, which patients, what mechanism? Circulation 2006;113:
1553-5.
44. Goto S. Cilostazol: potential mechanism of action for antithrombotic
effects accompanied by a low rate of bleeding. Atheroscler Suppl
2005;6:3-11.
45. Serebruany VL, Glassman AH, Malinin AI, et al. Platelet/
endothelial biomarkers in depressed patients treated with the
selective serotonin reuptake inhibitor sertraline after acute
coronary events: the Sertraline AntiDepressant Heart Attack
Randomized Trial (SADHART) Platelet Substudy. Circulation
2003;108:939-44.
American Heart Journal
Volume 162, Number 1
Berger et al 124.e1

Appendix. Supplementary data

Supplemental Figure 6

Association between the RR of aspirin versus placebo or control on the

prevention of MCE and year of study publication (a surrogate for
changes in background cardiovascular preventive therapies) (A),
baseline cardiovascular risk (as assessed by the annual risk of MCE on
placebo) (B), mean age of trial participant (C), percent female of trial
enrollment (D), and dose of aspirin used (E). Major cardiovascular
events defined as the composite of nonfatal MI, nonfatal stroke, or
cardiovascular death. CV, Cardiovascular.
 American Heart Journal
124.e2 Berger et al July 2011

Supplemental Figure 7

Association between the RR of aspirin versus placebo or control on the

risk of major bleeding and year of study publication (a surrogate for
changes in background cardiovascular preventive therapies) (A),
baseline cardiovascular risk (as assessed by the annual risk of MCE on
placebo) (B), mean age of trial participant (C), percent female of trial
enrollment (D), and dose of aspirin used (E). Major cardiovascular
events defined as the composite of nonfatal MI, nonfatal stroke, or
cardiovascular death. CV, Cardiovascular.