Anti Platelet Drugs Accp 8th Ed

Antiplatelet Drugs* : American College of
Chest Physicians Evidence-Based Clinical

Practice Guidelines (8th Edition)
Carlo Patrono, Colin Baigent, Jack Hirsh and Gerald Roth
Chest 2008;133;199S-233S
DOI 10.1378/chest.08-0672
The online version of this article, along with updated information

and services can be found online on the World Wide Web at:
http://chestjournal.chestpubs.org/content/133/6_suppl/199S.f
ull.html
CHEST is the official journal of the American College of Chest

Physicians. It has been published monthly since 1935. Copyright 2008
by the American College of Chest Physicians, 3300 Dundee Road,
Northbrook, IL 60062. All rights reserved. No part of this article or PDF
may be reproduced or distributed without the prior written permission
of the copyright holder.
(http://chestjournal.chestpubs.org/site/misc/reprints.xhtml)
ISSN:0012-3692
Downloaded from chestjournal.chestpubs.org by guest on March 22, 2010

© 2008 American College of Chest Physicians
Supplement
ANTITHROMBOTIC AND THROMBOLYTIC THERAPY 8TH ED: ACCP GUIDELINES
Antiplatelet Drugs*
American College of Chest Physicians
Evidence-Based Clinical Practice Guidelines
(8th Edition)
Carlo Patrono, MD; Colin Baigent, MD; Jack Hirsh, MD, FCCP;
and Gerald Roth, MD
This article about currently available antiplatelet drugs is part of the Antithrombotic and Thrombo-
lytic Therapy: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines
(8th Edition). It describes the mechanism of action, pharmacokinetics, and pharmacodynamics of
aspirin, reversible cyclooxygenase inhibitors, thienopyridines, and integrin ␣IIb␤3 receptor antago-
nists. The relationships among dose, efficacy, and safety are thoroughly discussed, with a mechanistic
overview of randomized clinical trials. The article does not provide specific management recommen-
dations; however, it does highlight important practical aspects related to antiplatelet therapy,
including the optimal dose of aspirin, the variable balance of benefits and hazards in different clinical
settings, and the issue of interindividual variability in response to antiplatelet drugs.
(CHEST 2008; 133:199S–233S)
Key words: abciximab; antiplatelet drugs; aspirin; clopidogrel; dipyridamole; eptifibatide; platelet pharmacology; resistance;
ticlopidine; tirofiban
Abbreviations: ACE ⫽ angiotensin-converting enzyme; ADP ⫽ adenosine diphosphate; AMP ⫽ adenosine monophos-
phate; ATT ⫽ Antithrombotic Trialists; CAPRIE ⫽ Clopidogrel vs Aspirin in Patients at Risk of Ischemic Events;
CHD ⫽ coronary heart disease; CI ⫽ confidence interval; COMMIT ⫽ Clopidogrel and Metoprolol Myocardial Infarction
Trial; COX ⫽ cyclooxygenase; CURE ⫽ Clopidogrel in Unstable Angina to Prevent Recurrent Events; EPIC ⫽ Evaluation of
7E3 for the Prevention of Ischemic Complications; ESPS ⫽ European Stroke Prevention Study; ESPRIT ⫽ European Stroke
Prevention Reversible Ischemia Trial; FDA ⫽ Food and Drug Administration; GP ⫽ glycoprotein; INR ⫽ international
normalized ratio; MI ⫽ myocardial infarction; NSAID ⫽ nonsteroidal antiinflammatory drug; OR ⫽ odds ratio;
PCI ⫽ percutaneous coronary intervention; PE ⫽ pulmonary embolism; PG ⫽ prostaglandin; PTCA ⫽ percutaneous trans-
luminal coronary angioplasty; RR ⫽ rate ratio; TIA ⫽ transient ischemic attack; TX ⫽ thromboxane; TTP ⫽ thrombotic
thrombocytopenic purpura
P latelets are vital components of normal hemosta-

sis and key participants in atherothrombosis by
though platelet adhesion and activation can be
viewed as a physiologic repair response to the sud-
virtue of their capacity to adhere to injured blood den fissuring or rupture of an atherosclerotic plaque,
vessels and to accumulate at sites of injury.1 Al- uncontrolled progression of such a process through a
series of self-sustaining amplification loops can lead
*From the Catholic University School of Medicine (Dr. Patrono), to intraluminal thrombus formation, vascular occlu-
Rome, Italy; Clinical Trial Service Unit (Dr. Baigent), University sion, and transient ischemia or infarction. Currently
of Oxford, Oxford, UK; Hamilton Civic Hospitals (Dr. Hirsh),
Henderson Research Centre, Hamilton, ON, Canada; and Seattle available antiplatelet drugs interfere with some steps
VA Medical Center (Dr. Roth), Seattle, WA. in the activation process, including adhesion, release,
Dr. Patrono was supported in part by a grant from the European and/or aggregation,1 and have a measurable impact
Commission FP6 (LSHM-CT-2004-005033).
Manauscript accepted December 20, 2007. on the risk of arterial thrombosis that cannot be
Reproduction of this article is prohibited without written permission dissociated from an increased risk of bleeding.2
from the American College of Chest Physicians (www.chestjournal.
org/misc/reprints.shtml).
In discussing antiplatelet drugs, it is important to
Correspondence to: Carlo Patrono, MD, Catholic University appreciate that approximately 1011 platelets are pro-
School of Medicine, Largo F. Vito 1, 00168 Rome, Italy; e-mail: duced each day under physiologic circumstances, a
carlo.patrono@rm.unicatt.it
DOI: 10.1378/chest.08-0672 level of production that can increase up to 10-fold at
www.chestjournal.org CHEST / 133 / 6 / JUNE, 2008 SUPPLEMENT 199S

times of increased need.3 Platelets are anucleate (PGI2), nitric oxide, CD39/ecto-ADPase, and platelet
blood cells that form by fragmentation of megakaryo- endothelial cell adhesion molecule-1. Some drugs may
cyte cytoplasm and have a maximum circulating life interfere with these regulatory pathways, as exempli-
span of about 10 days in humans.3 Platelets provide fied by the dose-dependent inhibition of PGI2 produc-
a circulating source of chemokines, cytokines, and tion by aspirin and other COX inhibitors.2
growth factors, which are preformed and packaged
in storage granules. Moreover, activated platelets can 2.0 Aspirin and Other COX Inhibitors
synthesize prostanoids (primarily, thromboxane [TX]
A2) from arachidonic acid released from membrane Aspirin has been thoroughly evaluated as an anti-
phospholipids through rapid coordinated activation platelet drug6 and was found to prevent vascular
of phospholipase(s), cyclooxygenase (COX)-1 and TX death by approximately 15% and nonfatal vascular
synthase (Fig 1). Newly formed platelets also express events by about 30% in a metaanalysis of ⬎ 100
the inducible isoforms of COX (COX-2) and prosta- randomized trials in high-risk patients.7
glandin (PG) E synthase, and this phenomenon is
markedly amplified in association with accelerated 2.1 Mechanism of Action of Aspirin
platelet regeneration.4 Although activated platelets The best characterized mechanism of action of the
are not thought to synthesize proteins de novo, they drug is related to its capacity to inactivate permanently
can translate constitutive messenger RNAs into pro- the COX activity of prostaglandin H-synthase-1 and -2
teins, including interleukin-1␤, over several hours.5 (also referred to as COX-1 and COX-2).8 –12 These
Thus, platelets may play previously unrecognized isozymes catalyze the first committed step in prosta-
roles in inflammation and vascular injury, and anti- noid biosynthesis (ie, the conversion of arachidonic
platelet strategies may be expected to affect platelet- acid to PGH2) [Fig 1]. PGH2 is the immediate
derived protein signals for inflammatory and/or pro- precursor of PGD2, PGE2, PGF2␣, PGI2, and TXA2.
liferative responses.1 COX-1 and COX-2 are homodimers of a ⬃ 72 kd
Negative modulation of platelet adhesion and aggre- monomeric unit. Each dimer has three independent
gation is exerted by a variety of physiologic mecha- folding units: an epidermal growth factor-like do-
nisms, including endothelium-derived prostacyclin main; a membrane-binding domain; and an enzy-
Figure 1. Arachidonic acid metabolism and mechanism of action of aspirin. Arachidonic acid, a
20-carbon fatty acid containing four double bonds, is liberated from the sn2 position in membrane
phospholipids by several forms of phospholipase, which are activated by diverse stimuli. Arachidonic
acid is converted by cytosolic PGH synthases, which have both COX and hydroperoxidase activity, to
the unstable intermediate PGH2. The synthases are colloquially termed COXs and exist in two forms,
COX-1 and COX-2. Low-dose aspirin selectively inhibits COX-1, and high-dose aspirin inhibits both
COX-1 and COX-2. PGH2 is converted by tissue-specific isomerases to multiple prostanoids. These
bioactive lipids activate specific cell membrane receptors of the superfamily of G-protein-coupled
receptors. DP ⫽ PGD2 receptor; EP ⫽ PGE2 receptor; FP ⫽ PGF2␣ receptor; IP ⫽ prostacyclin
receptor; TP ⫽ TX receptor.
200S Antithrombotic and Thrombolytic Therapy 8th Ed: ACCP Guidelines

matic domain.12 Within the enzymatic domain, there COX-2 inhibitors20 that also supports the concept of
is the peroxidase catalytic site and a separate, but PGI2 acting as an important mechanism of throm-
adjacent site for COX activity at the apex of a narrow, boresistance in the setting of inadequate inhibition
hydrophobic channel. of platelet TXA2 biosynthesis.21
The molecular mechanism of permanent inactiva-
tion of COX activity by aspirin is related to blockade 2.2 Pharmacokinetics
of the COX channel as a consequence of acetylation
Aspirin is rapidly absorbed in the stomach and
of a strategically located serine residue (Ser529 in
upper intestine. Peak plasma levels occur 30 to 40
the human COX-1, Ser516 in the human COX-2)
min after aspirin ingestion, and inhibition of plate-
that prevents access of the substrate to the catalytic
let function is evident by 1 h. In contrast, it can
site of the enzyme.13 The hydrophobic environment
take up to 3 to 4 h to reach peak plasma levels after
of the COX channel stabilizes the modified serine
administration of enteric-coated aspirin. If only
side-chain against hydrolysis.13 Thus, inhibition of
enteric-coated tablets are available, and a rapid
COX-1– dependent platelet function can be achieved
effect is required, the tablets should be chewed. The
with low doses of aspirin given once daily. In con-
oral bioavailability of regular aspirin tablets is ap-
trast, inhibition of COX-2– dependent pathophysio-
proximately 40 to 50% over a wide range of doses.22
logic processes (eg, hyperalgesia and inflammation)
A considerably lower bioavailability has been re-
requires larger doses of aspirin (probably because
ported for enteric-coated tablets and sustained-
acetylation is determined by the oxidative state of the
release, microencapsulated preparations.22 Lower
enzyme and is inhibited in cells with high peroxide
bioavailability of some enteric-coated preparations
tone)14 and a much shorter dosing interval (because
and poor absorption from the higher pH environ-
nucleated cells rapidly resynthesize the enzyme). Thus,
ment of the small intestine may result in inadequate
there is an approximately 100-fold variation in daily
platelet inhibition, particularly in heavier subjects.23
doses of aspirin when used as an antiinflammatory
Both a controlled-release formulation15 and a trans-
rather than as an antiplatelet agent. Furthermore,
dermal patch24 with negligible systemic bioavailabil-
the benefit/risk profile of the drug depends on the
ity have been developed in an attempt to achieve
dose and indication because its GI toxicity is dose
selective inhibition of platelet TXA2 production with-
dependent (see below).
out suppressing systemic PGI2 synthesis. The former
Human platelets and vascular endothelial cells
was used successfully in the Thrombosis Prevention
process PGH2 to produce primarily TXA2 and PGI2,
Trial (see below), but it remains unknown whether
respectively.11 TXA2 induces platelet aggregation
there is any advantage to the controlled-release
and vasoconstriction, whereas PGI2 inhibits platelet
formulation vis-à-vis plain aspirin.
aggregation and induces vasodilation.11 Whereas
The plasma concentration of aspirin decays with a
TXA2 is largely a COX-1– derived product (mostly
half-life of 15 to 20 min. Despite the rapid clearance
from platelets) and thus highly sensitive to aspirin
of aspirin from the circulation, the platelet-inhibitory
inhibition, vascular PGI2 can derive both from
effect lasts for the life span of the platelet25 because
COX-1 and, to a greater extent even under physio-
aspirin irreversibly inactivates platelet COX-1.8,9 As-
logic conditions, from COX-2.16 COX-1– dependent
pirin also acetylates the enzyme in megakaryocytes
PGI2 production occurs transiently in response to
before new platelets are released into the circula-
agonist stimulation (eg, bradykinin)15 and is sensitive
tion.10,26 –28 The mean life span of human platelets is
to aspirin inhibition. COX-2–mediated PGI2 produc-
approximately 8 to 10 days. Therefore, about 10 to
tion occurs long term in response to laminar shear
12% of circulating platelets are replaced every
stress17 and is largely insensitive to aspirin inhibition
24 h.29,30 However, the recovery of TXA2 biosynthe-
at conventional antiplatelet doses. This may explain
sis in vivo following prolonged aspirin administration
the substantial residual COX-2– dependent PGI2
is somewhat faster than predicted by the rate of
biosynthesis in vivo at daily doses of aspirin in the
platelet turnover,18 possibly because of the nonlinear
range of 30 to 100 mg,18 despite transient suppres-
relationship between inhibition of platelet COX-1
sion of COX-1– dependent PGI2 release.15 It is not
activity and inhibition of TXA2 biosynthesis in vivo31
established that more profound suppression of PGI2
(Fig 2).
formation by higher doses of aspirin is sufficient to
initiate or predispose to thrombosis. However, two
2.3 Issues Concerning the Antithrombotic Effects of
lines of evidence suggest that PGI2 is thrombopro-
Aspirin
tective. The first is the observation that mice lacking
the PGI2 receptor had increased susceptibility to A number of issues related to the clinical efficacy
experimental thrombosis.19 The second is the obser- of aspirin continue to be debated. These include the
vation of the cardiovascular toxicity associated with following: (1) the optimal dose of aspirin in order to

TXB2 TXB2 Pharmacologic
Production Production Inhibition
Ex Vivo In Vivo Ex Vivo vs In Vivo
Platelet TXA2 Extra Platelet
Synthesis Sources
H2O
Whole Blood 100
Percentage Inhibition of Urinary 2,3-

300 Clotting at 37°C
TXB2
dinor-TXB2 Excretion In Vivo

Serum TXB2 ng/ml
Liver 80
Enzymes
200 60
2,3-dinor-TXB2
& other 40
100 metabolites
Maximal Biosynthetic
Capacity 300-400 20
ng/ml in 1 hr
0 Urine 0
0 30 60 min 0 20 40 60 80 100
Time Calculated Rate of TXB2 Percentage Inhibition of Serum
Production In Vivo: TXB2 Ex Vivo
0.1 ng/kg/min
Figure 2. Maximal capacity of human platelets to synthesize TXB2, rate of TXB2 production in healthy
subjects, and relationship between the inhibition of platelet COX activity and TXB2 biosynthesis in
vivo. Left panel: The level of TXB2 production stimulated by endogenous thrombin during whole-blood
clotting at 37°C.62 Center panel: The metabolic fate of TXA2 in vivo and the calculated rate of its
production in healthy subjects on the basis of TXB2 infusions and measurement of its major urinary
metabolite. Right panel: The nonlinear relationship between inhibition of serum TXB2 measured ex
vivo and the reduction in the excretion of TX metabolite measured in vivo.31
maximize efficacy and minimize toxicity; (2) the The lowest effective dose of aspirin for these various
suggestion that part of the antithrombotic effect of indications is shown in Table 1.
aspirin is unrelated to inhibition of platelet TXA2; The clinical effects of different doses of aspirin
and (3) the possibility that some patients may be have been compared directly in a relatively small
aspirin “resistant.” number of randomized trials.38 – 43 In the United
Kingdom TIA study,41 no difference in efficacy was
2.3.1 The Optimal Dose of Aspirin: Well-designed, found between 300 and 1,200 mg/d of aspirin (see
placebo-controlled randomized trials have shown below). In a study of 3,131 patients after a TIA or
that aspirin is an effective antithrombotic agent when minor ischemic stroke, aspirin in a dose of 30 mg/d
used long term in doses ranging from 50 to 100 mg/d, was compared with a dose of 283 mg/d, and the
and there is a suggestion that it is effective in doses
as low as 30 mg/d.6,7 Aspirin, 75 mg/d, was shown to
be effective in reducing the risk of acute myocardial Table 1—Vascular Disorders for Which Aspirin Has
infarction (MI) or death in patients with unstable Been Shown To Be Effective and Lowest Effective Dose
angina32 and chronic stable angina,33 as well as in (Section 2.3.1)
reducing stroke or death in patients with transient Lowest Effective
cerebral ischemia34 and the risk of postoperative Disorder Daily Dose, mg
stroke after carotid endarterectomy.35 In the Euro- TIA and ischemic stroke* 50
pean Stroke Prevention Study (ESPS)-2, aspirin 25 Men at high cardiovascular risk 75
mg bid was effective in reducing the risks of stroke Hypertension 75
and of the composite outcome stroke or death in Stable angina 75
Unstable angina* 75
patients with prior stroke or transient ischemic attack
Severe carotid artery stenosis* 75
(TIA).36 Moreover, in the European Collaboration Polycythemia vera 100
on Low-Dose Aspirin in Polycythemia vera Trial,37 Acute MI 160
aspirin, 100 mg/d, was effective in preventing throm- Acute ischemic stroke* 160
botic complications in patients with polycythemia *Higher doses have been tested in other trials and were not found to
vera, despite a higher-than-normal platelet count. confer any greater risk reduction.

hazard ratio for the group receiving the lower dose (35-day) mortality as well as nonfatal reinfarction and
was 0.91 (95% confidence interval [CI], 0.76 to stroke55; transient cerebral ischemia in which doses
1.09).42 The Acetylsalicylic Acid and Carotid Endar- between 50 and 1,200 mg/d were effective34,36,41,56 –58;
terectomy Trial reported that the risk of stroke, MI, acute ischemic stroke in which doses of 160 to 300
or death within 3 months of carotid endarterectomy mg/d were effective in reducing early mortality and
is significantly lower for patients taking 81 or 325 stroke recurrence59,60; and polycythemia vera in
mg/d aspirin than for those taking 650 or 1,300 mg which 100 mg,37 but not 900 mg,61 was effective in
(6.2% vs 8.4%; p ⫽ 0.03).43 Thus, there is no con- reducing fatal and nonfatal vascular events.
vincing evidence from randomized studies that have Thus, aspirin is an effective antithrombotic agent
compared different doses of aspirin that higher doses in doses between 50 and 1,500 mg/d. It is also
are more effective in reducing the risk of serious possible from the results of the Dutch TIA study that
vascular events. In fact, both this limited set of 30 mg/d is effective.42 There is no evidence that low
randomized comparisons and the indirect compari- doses (50 to 100 mg/d) are less effective than high
sons reported in the overview of the Antithrombotic doses (650 to 1,500 mg/d) and, in fact, the opposite
Trialists’ (ATT) Collaboration (Table 2) are compat- may be true. These clinical findings are consistent
ible with the reverse (ie, blunting of the antithrom- with saturability of platelet COX-1 inactivation at
botic effect at higher doses of aspirin, consistent with doses as low as 30 mg/d.62
dose-dependent inhibition of PGI2). Such inhibition There is evidence, however, that doses of approx-
of PGI2 may be a potential mechanism by which imately 300 mg/d produce fewer GI side effects than
COX-2 inhibitors produce an excess risk of MI (see doses of approximately 1,200 mg/d.41 There is also
below). some evidence that a dose of 30 mg/d produces
The antithrombotic effects of a range of doses of fewer side effects than 300 mg/d.42 The Clopidogrel
aspirin also have been compared with an untreated in Unstable Angina To Prevent Recurrent Events
control group in a number of thrombotic vascular (CURE) investigators have retrospectively investi-
disorders. The doses have varied between 50 and gated the relationship between the aspirin dose (the
1,500 mg/d. Aspirin has been shown to be effective CURE protocol recommended 75 to 325 mg/d) and
in the following conditions: unstable angina in which risk of major bleeding.63 This study was a random-
the incidence of acute MI or death was significantly ized comparison of clopidogrel with placebo on a
reduced in four separate studies using daily doses of “background” of aspirin therapy. Patients with acute
75 mg,32 325 mg,44 650 mg,45 and 1,300 mg46; stable coronary syndromes receiving aspirin, ⱕ 100 mg/d,
angina in which a dose of 75 mg/d reduced the had the lowest rate of major or life-threatening
incidence of acute MI or sudden death33; aortocoro- bleeding complications both in the placebo (1.9%)
nary bypass surgery in which the incidence of early and in the clopidogrel (3%) arms of the trial. Bleed-
occlusion was similarly reduced with daily doses of ing risks increased with increasing aspirin dose with
100 mg,47 325 mg,48 975 mg,49 and 1,200 mg49; or without clopidogrel.63
thromboprophylaxis of patients with prosthetic heart In summary, the saturability of the antiplatelet
valves who also received warfarin in whom the effect of aspirin at low doses, the lack of dose-
incidence of systemic embolism was reduced with response relationship in clinical studies evaluating its
daily doses of 100 mg,50 500 mg,51 and 1,500 mg52,53; antithrombotic effects, and the dose dependence of
thromboprophylaxis of patients with arterial venous its side effects all support the use of as low a dose of
shunts undergoing long-term hemodialysis in whom aspirin as has been found to be effective in the
a dose of 160 mg/d was shown to be effective54; acute treatment of various thromboembolic disorders (Ta-
MI in which a dose of 162.5 mg/d reduced early ble 1). Use of the lowest effective dose of aspirin (50
to 100 mg/d for long-term treatment) is currently the
most appropriate strategy to maximize its efficacy
and minimize its toxicity.6
Table 2—Indirect Comparison of Aspirin Doses
Reducing Vascular Events in High-Risk 2.3.2 Effects of Aspirin Not Related to TXA2:
Patients (Section 2.3.1)*
Aspirin has been reported to have effects on hemostasis
Aspirin Dose, No. of No. of Odds that are unrelated to its ability to inactivate platelet
mg/d Trials Patients Reduction COX-1. These include dose-dependent inhibition of
500–1,500 34 22,451 19 ⫾ 3% platelet function,64 – 68 enhancement of fibrinolysis,69 –71
160–325 19 26,513 26 ⫾ 3% and suppression of plasma coagulation.72–75
75–150 12 6,776 32 ⫾ 6% In contrast to the saturable and well-characterized
⬍ 75 3 3,655 13 ⫾ 8% (nanomolar aspirin concentration, rapid time course,
*Data are from Lindemann et al5/2001. physiologic conditions, single serine modification)

inhibition of COX-1 by aspirin,13,62,76 the putative Symptomatic Carotid Endarterectomy Trial study79
mechanisms underpinning the non-PG effects of suggested that aspirin in doses of ⱖ 650 mg/d might
aspirin on hemostasis are dose dependent and less be more effective than ⱕ 325 mg/d for the preven-
clearly defined. For example, inhibition of shear- tion of perioperative stroke in patients having carotid
induced platelet aggregation depends on the level of artery surgery.80 This retrospective observation was
aspirin provided, and enhanced fibrinolysis due to refuted by a second prospective study, the Acetylsal-
N-acetylation of lysyl residues of fibrinogen is seen in icylic Acid and Carotid Endarterectomy Trial,43
vivo with high doses of aspirin (650 mg bid)69 and which tested the hypothesis that the wide area of
proceeds more rapidly in vitro under nonphysiologic collagen exposed by endarterectomy is a sufficiently
alkaline conditions.77 Aspirin suppresses plasma co- strong stimulus to platelet aggregation to require a
agulation through several mechanisms. The first, larger dose of aspirin. Approximately 3,000 patients
initially described in 1943 by Link et al and con- scheduled for carotid endarterectomy were ran-
firmed by others,72,73 is caused by an antivitamin K domly assigned 81, 325, 650, or 1,300 mg/d aspirin,
effect of aspirin. It requires very high doses of aspirin started before surgery and continued for 3 months.
and does not contribute to the antithrombotic effect The combined rate of stroke, MI, or death at 3
of aspirin when the drug is used in doses up to 1,500 months was significantly (p ⫽ 0.03) lower in the
mg/d. The second is platelet dependent and is low-dose groups (6.2%) than in the high-dose groups
characterized by inhibition of thrombin generation (8.4%) [primary analysis]. There were no significant
in a whole blood system.74,75 A single dose of 500 mg differences between the 81-mg and 325-mg groups
depresses the rate of thrombin generation, whereas or between the 650-mg and 1,300-mg groups in any
repeated daily dosing with 300 mg of aspirin reduces of the secondary analyses of the data.43
the total amount of thrombin formed.78 An interac- A subgroup analysis of the Physicians’ Health
tion with platelet phospholipids, which is blunted in Study,81 based on post hoc measurements of baseline
hypercholesterolemia, has been proposed to explain plasma C-reactive protein performed in 543 appar-
the effects of aspirin on thrombin generation.78 It is ently healthy men who subsequently had MI, stroke,
possible (but unproven) that this effect occurs as a or venous thrombosis, and in 543 study participants
consequence of impaired platelet coagulant activity who did not report vascular complications, has found
secondary to inhibition of TX-dependent platelet that the reduction in the risk of a first MI associ-
aggregation. It is unknown whether lower doses of ated with the use of aspirin (325 mg on alternate
aspirin are able to produce this effect. This sort of in days) appears to be directly related to the level of
vitro effect has been shown for other platelet inhibi- C-reactive protein, raising the possibility of antiin-
tors, such as glycoprotein (GP)-IIb/IIIa antagonists (see flammatory as well as antiplatelet effects of the drug
below). Furthermore, high-dose aspirin can cause ab- in cardiovascular prophylaxis.82 This hypothesis is
normal coagulation in vitro by direct acetylation of one unlikely to be correct because, as noted above, the
or more clotting factors. This can be demonstrated in antiinflammatory effects of aspirin and other nonste-
platelet-poor plasma and, thus, is not related to platelet roidal antiinflammatory drugs (NSAIDs) are largely
inhibition or vitamin K antagonism. related to their capacity to inhibit COX-2 activity
Additional studies in both animal models and induced in response to inflammatory cytokines,12 as
human subjects have reported antithrombotic effects these clinical effects can be fully reproduced by
of aspirin that may occur, at least in part, through highly selective COX-2 inhibitors (coxibs) in patients
mechanisms unrelated to inactivation of platelet with rheumatoid arthritis.83 As shown in Table 3, the
COX-1. In animal models, Buchanan et al66 and dose and time dependence of the effects of aspirin
Hanson et al64 reported that optimal antithrombotic on nucleated inflammatory cells expressing COX-2
activity of aspirin required doses in excess of those vs anucleated platelets expressing COX-1 are mark-
required to inhibit TXA2. In clinical studies, the edly different, thus making a clinically relevant anti-
results of a subgroup analysis of the North American inflammatory effect of the drug at 325 mg every
Table 3—Dose and Time Dependence of the Effects of Aspirin on Platelets and Inflammatory Cells (Section 2.3.2)
Single Duration of Prostanoid Cumulative Effects Upon Daily

Cellular Target Enzyme Dose,* mg Suppression, h Repeated Dosing Dose,† mg
Platelets COX-1 100 24–48 Yes 50–81

Inflammatory cells COX-2 ⱖ 650 3–4 No 3,000–5,000
*Dose causing full suppression of prostanoid formation and/or clinically detectable functional effect after single dosing.
†Range of doses shown clinically effective in long-term trials of cardiovascular protection or rheumatoid arthritis.

other day pharmacologically implausible. Finally, issue. The fact that some patients may experience
aspirin has been reported to modify the way in which recurrent vascular events despite long-term aspirin
neutrophils and platelets84 or erythrocytes and plate- therapy should be properly labeled as treatment
lets85,86 interact, to protect endothelial cells from failure rather than aspirin resistance. Treatment
oxidative stress,87 and to improve endothelial dys- failure is a common phenomenon occurring with all
function in atherosclerotic patients.88 However, nei- drugs (eg, lipid-lowering or antihypertensive drugs).
ther the molecular mechanism(s) nor the dose de- Given the multifactorial nature of atherothrombosis
pendence of these effects have been clearly and the possibility that platelet-mediated thrombosis
established. Although improved endothelial dysfunc- may not be responsible for all vascular events, it is
tion could reflect an antiinflammatory effect of not surprising that only a fraction (usually one fourth
aspirin of relevance to atherogenesis, it should be to one third) of all vascular complications can be
emphasized that the hypothesis has never been prevented by any single preventive strategy.
tested by an appropriately sized controlled prospec- It has been reported that a variable proportion (up
tive study. to one fourth) of patients with cerebrovascular dis-
All of the evidence detailed above suggesting ease only achieve partial inhibition of platelet aggre-
dose-dependent effects for aspirin is indirect and gation at initial testing, and some (up to one third)
inconsistent with the failure to show a dose effect in seem to develop resistance to aspirin over time, even
randomized clinical trials and in the ATT overview with increasing doses.91–93 The results of these long-
analysis.7 This failure to show a dose effect is the term studies carried out by Helgason et al are at
critical point of the discussion because it correlates variance with those of a short-term study of Weksler
with the saturability of the aspirin effect on platelet et al,94 showing that aspirin, 40 mg/d, inhibited
COX-1. For example, in studies with purified en- platelet aggregation and TXA2 formation as effec-
zyme and with isolated platelets, nanomolar concen- tively as higher doses of aspirin in patients who had
trations of aspirin will completely block PG synthesis recent cerebral ischemia. Variable platelet responses
within 20 min after exposure.89 Higher concentra- to aspirin have also been described in patients with
tions and longer exposures will not alter the inhibi- peripheral arterial disease95 and with ischemic heart
tory effect of aspirin on PG synthesis because of this disease.96 –98 In the Buchanan and Brister study,96
saturable quality. Exactly the same feature (maximal aspirin nonresponders were identified on the basis of
effect at low doses, absence of dose effect) is seen in bleeding time measurements. Approximately 40% of
clinical trials with aspirin as an antithrombotic agent. patients undergoing elective coronary artery bypass
When one raises the dose of aspirin in this situation, grafting showed no prolongation of bleeding time in
no further or additional effect can be appreciated response to aspirin. This finding was associated with
because the critical event has already taken place, increased platelet adhesion and 12-HETE synthe-
namely, maximal inhibition of platelet TX synthesis. sis.96 In contrast, repeated measurements of platelet
Thus, the consistency of dose requirements and aggregation performed over 24 months of placebo-
saturability of the effects of aspirin in acetylating the controlled treatment by Berglund and Wallentin99
platelet enzyme,8 inhibiting TXA2 production,25,62 demonstrated that 100 patients with unstable coro-
and preventing atherothrombotic complications6,7 nary artery disease randomized to receive aspirin,
constitutes the best evidence that aspirin prevents 75 mg/d, in the Research Group on Instability in
thrombosis through inhibition of TXA2 production. Coronary Artery Disease in Southeast Sweden
It is likely, therefore, that any of the potential effects study32 had consistently reduced platelet aggregation
of aspirin on other determinants of arterial throm- without attenuation during long-term treatment.
bosis are much less important than the inhibition of Based on measurements of platelet aggregation in
platelet COX-1 activity. response to arachidonate and adenosine diphosphate
(ADP), 5% and 24% of patients with stable cardio-
2.3.3 Aspirin “Resistance”: The term aspirin resis- vascular disease who were receiving aspirin (325
tance has been used to describe a number of differ- mg/d for ⱖ 7 days) were defined as resistant and
ent phenomena, including the inability of aspirin to semiresponders, respectively.97 Using a variety of
(1) protect individuals from thrombotic complica- techniques, including conventional aggregometry,
tions, (2) cause a prolongation of the bleeding time, shear stress-induced activation, and the expression of
(3) reduce TXA2 production, or (4) produce a typical platelet surface receptors, Sane et al98 recently re-
effect on one or more in vitro tests of platelet ported that 57% of a group of 88 patients with
function.90 From a therapeutic standpoint, it is im- documented heart failure who had been treated with
portant to establish whether aspirin resistance can be aspirin, 325 mg/d, for ⱖ 1 month showed aspirin
overcome by increasing the dose of aspirin, but nonresponsiveness. Overall, the majority of these
unfortunately very few data bear directly on this studies were characterized by the following major

limitations: (1) biochemical or witnessed verification cally expect if aspirin resistance could be overcome at
of patient’s adherence to the prescribed therapy least in part by increasing the daily dose of the drug.
were absent; (2) there was a single measurement of Gum et al106 reported that 5% of 326 stable
any given test; (3) intrasubject and intersubject cardiovascular patients were aspirin resistant based
variability and stability of the assay over time were on the results of platelet aggregation induced by
usually not reported; (4) the criteria to define the ADP and arachidonic acid. The aspirin-resistant
normal vs the aspirin-resistant range and the assay group had an increased risk of death, MI, or cere-
conditions differed among studies; (5) doses of aspi- brovascular accident during almost 2 years of follow-up.
rin were heterogeneous, ranging from 75 to 1,300 There were, however, relatively few events in this
mg; and (6) none of these studies were properly study, and the rationale for the particular definition
controlled. of aspirin resistance is uncertain.
Among a wide range of patients with vascular
Lack of biochemical assessment of compliance is a
disease in whom the annual rate of serious vascular
major issue for the majority of studies assessing
events ranges from 40 to 80 per 1,000, aspirin
platelet function in response to aspirin, and this
typically prevents at least 10 to 20 fatal and nonfatal
aspect is crucial in studies investigating aspirin vascular events for every 1,000 patients treated for 1
unresponsiveness. Interestingly, a recent study100 year.6 Thus, 30 to 60 vascular events are expected to
in 190 patients with a history of MI compared occur for every 1,000 patients treated with low-dose
arachidonate-induced platelet aggregation in pa- aspirin for 1 year not because of resistance, but
tients while receiving their usual aspirin therapy, because of the multifactorial nature of atherothrom-
after 7 days of withdrawal, and 24 h after a single bosis. Thus, we do not agree with the definition
witnessed intake of aspirin of 325 mg. Although 9% given by Wang et al,107 whereby “in its broadest
of the patients who declared having taken their usual sense, resistance refers to the continued occurrence
therapy failed to show inhibition of platelet aggrega- of ischemic events despite adequate antiplatelet
tion, this percentage dropped to ⬍ 1% (1 patient of therapy and compliance.” Indeed, as suggested by
190) after a witnessed dose.100 Furthermore, this Hennekens et al,108 “given the multiple pathways by
single patient admitted NSAID intake 12 h before which platelets may be activated, it is perhaps more
testing. Similar results were reported in the study by surprising that a clinical benefit is detectable in
Lev et al,101 where after a witnessed dose of 325 mg randomized trials of cardiovascular disease than that
of aspirin, the mean of arachidonic acid-induced treatment failures complicate aspirin therapy.”
light transmission aggregometry became ⬍ 20% (the At least three potential mechanisms may underlie
established limit to define resistance) in formerly the occurrence of aspirin-resistant TXA2 biosynthe-
resistant patients. Other studies have reported up to sis. The transient expression of COX-2 in newly
40% noncompliance with long-term aspirin use.102 It formed platelets in clinical settings of enhanced
is therefore clear that questionnaires cannot be a platelet turnover4 is a potentially important mecha-
reliable parameter to assess the compliance to any nism that deserves further investigation. Extraplate-
given treatment, including aspirin, and that studies let sources of TXA2 (eg, monocyte/macrophage
not relying on salicylate measurements or serum COX-2) may contribute to aspirin-insensitive TXA2
TXB2 have a major, intrinsic bias, seriously hamper- biosynthesis in acute coronary syndromes.109 Fur-
ing the interpretation of results. Furthermore, the thermore, Catella-Lawson et al110 reported that con-
few studies directly comparing different functional comitant administration of a traditional NSAID (eg,
assays failed to find any significant agreement be- ibuprofen) may interfere with the irreversible inac-
tween tests, generating the disappointing conclusion tivation of platelet COX-1 by aspirin. This is due to
that aspirin nonresponsiveness may be highly test competition for a common docking site within the
specific. COX channel (arginine-120), which aspirin binds to
Several relatively small studies (n ⫽ 39 to 180) of with weak affinity before irreversible acetylation of
stroke patients103–105 have suggested that aspirin Serine-529.13 This pharmacodynamic interaction also
resistance may contribute to treatment failure (ie, has been described between naproxen and aspirin111
recurrent ischemic events while on antiplatelet ther- but does not occur with rofecoxib,110 celecoxib,112 or
apy) and that doses higher than 500 mg may be more diclofenac,110 drugs endowed with variable COX-2
effective than lower doses in limiting this phenome- selectivity.83 Thus, concomitant treatment with
non. The uncontrolled nature and small sample size readily available over-the-counter NSAIDs may limit
of these studies make it difficult to interpret the the cardioprotective effects of aspirin and contribute
results. As noted above, a much larger database failed to aspirin resistance. Based on current analysis of
to substantiate a dose-dependent effect of aspirin in available data,113–115 the US Food and Drug Admin-
stroke prevention,7 an effect that one would theoreti- istration (FDA) has issued a statement informing

patients and health-care professionals that ibuprofen and the absolute benefits of antiplatelet therapy are
can interfere with the antiplatelet effect of low-dose somewhat heterogeneous in different clinical settings.
aspirin (81 mg/d), potentially rendering aspirin less In the Second International Study of Infarct Sur-
effective when used for cardioprotection and stroke vival,55 a single tablet of aspirin (162.5 mg) started
prevention (http://www.fda.gov/cder/drug/infopage/ within 24 h of the onset of symptoms of a suspected
ibuprofen/default.htm). MI and continued daily for 5 weeks produced highly
The clinical relevance of aspirin-resistant TXA2 significant reductions in the risk of vascular mortality
biosynthesis has been explored by Eikelboom et (by 23%), nonfatal reinfarction (by 49%), and non-
al,116 who performed a nested case-control study fatal stroke (by 46%). There was no increase in
of baseline urinary TX metabolite excretion in hemorrhagic stroke or GI bleeding in the aspirin-
relation to the occurrence of major vascular events treated patients and only a small increase in minor
in aspirin-treated high-risk patients enrolled in the bleeding.55 Treatment of 1,000 patients with sus-
Heart Outcomes Prevention Evaluation trial. After pected acute MI with aspirin for 5 weeks will result
adjustment for baseline differences, the odds for in approximately 40 patients in whom a vascular
the composite outcome of MI, stroke, or cardio- event is prevented,7 with a proportional odds reduc-
vascular death increased with each increasing tion of 30% (see the “Acute ST-Segment Elevation
quartile of 11-dehydro-TXB2 excretion, with pa- Myocardial Infarction” chapter).
tients in the upper quartile having a 1.8-times Two separate trials with a similar protocol, the
higher risk than those in the lower quartile. One International Stroke Trial59 and the Chinese Acute
limitation in this study, however, was the inability Stroke Trial,60 tested the efficacy and safety of early
to differentiate between variable compliance in aspirin use in acute ischemic stroke. Approximately
taking aspirin as prescribed (or avoiding NSAIDs) and 40,000 patients were randomized within 48 h of the
variable occurrence of aspirin-resistant sources of onset of symptoms to 2 to 4 weeks of daily aspirin
TXA2 biosynthesis. therapy (300 mg and 160 mg, respectively) or pla-
Thus, in summary, both the mechanism(s) and cebo. An overview of the results of both trials
clinical relevance of aspirin resistance, as defined suggests an absolute benefit of 9 fewer deaths or
by platelet aggregation measurements, remain to nonfatal strokes per 1,000 patients in the first month
be established.117 Until its true nature and preva- of aspirin therapy.7 The proportional odds reduction
lence vis-à-vis other antiplatelet drugs are better in fatal or nonfatal vascular events is only 10% in this
defined, no test of platelet function is recom- setting. Although the background risk of hemorrhagic
mended to assess the antiplatelet effect of aspirin stroke was threefold higher in the Chinese Acute
in the individual patient.118,119 On the other hand, Stroke Trial than in the International Stroke Trial, the
additional studies on the mechanisms and clinical absolute increase in this risk associated with early use of
relevance of aspirin-resistant TXA2 biosynthesis aspirin was similar in the two studies (excess 2 per 1,000
are clearly warranted. patients).59,60 The broad clinical implications of these
findings are discussed in the “Antithrombotic and
2.4 The Antithrombotic Effect of Aspirin Thrombolytic Therapy for Ischemic Stroke” chapter. In
terms of their research implications, these results are
2.4.1 Prevention of Atherothrombosis in Different consistent with biochemical evidence of episodic plate-
Clinical Settings: The efficacy and safety of aspirin let activation during the first 48 h after the onset of
are documented from analysis of approximately 70 symptoms of an acute ischemic stroke and with sup-
randomized clinical trials that included ⬎ 115,000 pression of in vivo TXA2 biosynthesis in patients receiv-
patients at variable risk of thrombotic complications ing low-dose aspirin in this setting.120
of atherosclerosis. A detailed analysis of individual Long-term aspirin therapy confers conclusive net
trials is beyond the scope of this article. It is more benefit on risk of subsequent MI, stroke, or vascular
appropriately dealt within specific clinical sections of death among subjects with high risk of vascular
this volume. complications. These include patients with chronic
Aspirin has been tested in patients demonstrating stable angina,33 patients with prior MI,7 patients with
the whole spectrum of atherosclerosis, from appar- unstable angina,32,44 – 46 and patients with TIA or
ently healthy low-risk individuals to patients present- minor stroke34,36,41,56 –58 as well as other high-risk
ing with an acute MI or an acute ischemic stroke; categories.7 The proportional effects of long-term
similarly, trials have extended for as short as a few aspirin therapy on vascular events in these different
weeks’ duration or as long as 10 years.6,7 Although clinical settings are rather homogeneous, ranging
aspirin has been shown consistently to be effective in between 20% and 25% odds reduction based on an
preventing fatal and/or nonfatal vascular events in overview of all randomized trials.7 However, individ-
these trials, both the size of the proportional effects ual trial data show substantial heterogeneity, ranging

from no statistically significant benefits in patients vascular events. The question of whether aspirin is
with peripheral vascular disease to approximately effective for the primary prevention of vascular
50% risk reduction in patients with unstable angina.7 events has been addressed in a metaanalysis of
Although other factors may play a role, we interpret randomized trials.332
these findings as reflecting the variable importance Six primary prevention trials81,121–125 including
of TXA2 as a mechanism amplifying the hemostatic 92,873 participants were studied (Table 4). Mean
response to plaque destabilization in different clini- follow-up was approximately 6 years. There was a
cal settings. In terms of absolute benefit, these 15% reduction in the odds of cardiovascular events
protective effects of aspirin translate into avoidance (OR, 0.85; 95% CI, 0.79 to 0.92; p ⬍ 0.001) and
of a major vascular event in 50 per 1,000 patients highly significant reductions of 23% in total coronary
with unstable angina treated for 6 months and in 36 heart disease (CHD) [OR, 0.77; 95% CI, 0.70 to
per 1,000 patients with prior MI, stroke, or TIA 0.86; p ⬍ 0.001] and 24% in nonfatal MI (OR, 0.76;
treated for approximately 30 months.7 95% CI, 0.67 to 0.85; p ⬍ 0.001).
For patients with different manifestations of There was no overall effect on stroke (OR, 0.95;
ischemic heart or brain disease, a widespread 95% CI, 0.84 to 1.06; p ⫽ 0.3), but data were not
consensus exists in defining a rather narrow range available separately for hemorrhagic and nonhemor-
of recommended daily doses (ie, 75 to 160 mg) for rhagic stroke,126 –128 so the effects on these two
the prevention of MI, stroke, or vascular death. stroke subtypes could not be examined in detail.
This is supported by separate trial data in patients Aspirin had no significant effect on the aggregate of
randomized to treatment with low-dose aspirin or all vascular causes of death (OR, 0.89; 95% CI, 0.72
placebo as well as by an overview of all antiplatelet to 1.10; p ⫽ 0.3), or on overall mortality (OR, 0.94;
trials showing no obvious dose dependence, from 95% CI, 0.87 to 1.00; p ⫽ 0.07). In summary,
indirect comparisons, for the protective effects of therefore, in primary prevention, aspirin chiefly pre-
aspirin7 (Table 3). There is no convincing evidence vents nonfatal MI, and appears to have little effect
that the dose requirement for the antithrombotic on fatal vascular events.
effect of aspirin varies in different clinical settings.
Among most high-risk patient groups, the ex-
2.5 Balance of Benefit and Harm
pected number avoiding a serious vascular event by
using aspirin substantially exceeds the number expe- Previous metaanalyses of the effects of antiplatelet
riencing a major bleed. However, it is unclear therapy among people at high risk of occlusive
whether aspirin might benefit people who, although vascular disease7 have shown that the benefits of
apparently healthy, are at intermediate risk of serious aspirin far exceed the bleeding risks among such
Table 4 —Primary Prevention Trials (Section 2.4.1)
Mean Eligible Randomized

Year of Duration of Age Factorial Placebo
Trial Dates of Recruitment Publication Follow-up,* yr Target Population Range, yr Intervention Comparison Control
British 11/1/1978 to 11/1/1979 1988 5.6 Male physicians 19–90 500 mg/d None No
Doctors
Study
US physicians 8/24/1981 to 4/2/1984 1988 5.0 Male physicians 45–73 325 mg Beta-carotene Yes
alternate (alternate days)
days vs placebo
Thrombosis 2/6/1989 to 5/18/1994 1998 6.7 Men with risk 45–69 75 mg/d Warfarin vs placebo Yes
Prevention factors for
Trial CHD
Hypertension 10/12/1992 to 5/7/1994 1998 3.8 Men and women 31–75 75 mg/d Three target diastolic Yes
Optimal with diastolic BPs (⬍ 80 mm Hg,
Treatment BP 100–115 ⬍ 85 mm Hg,
mm Hg ⬍ 90 mm Hg)
Primary 6/8/1993 to 4/21/1998 2001 3.7 Men and women 45–94 100 mg/d Vitamin E vs open No
Prevention with one or control
Project more risk
factors for
CHD
Women’s 9/1992 to 5/1995 2005 10.1 Female health ⱖ 45 100 mg Vitamin E vs Yes
Health professionals alternate placebo
Study days
*Mean duration of follow-up among surviving participants within each trial.

patients. By contrast, the majority of participants 2.5.2 Deep Vein Thrombosis: The Pulmonary Em-
(92%) in the primary prevention trials were at low bolism Prevention Trial138 has established that aspi-
absolute risk of coronary disease; on average, the rin is effective in preventing venous thromboembo-
annual risk of a vascular event in the primary pre- lism after surgery for hip fracture. This was a
vention trials was only about one tenth of that double-blind multicenter study of 13,356 patients
occurring in the high-risk trials. Hence, although the undergoing surgery for hip fracture and of an addi-
proportional benefits of aspirin appeared broadly tional 4,088 patients undergoing elective hip or knee
similar in primary and secondary prevention, the arthroplasty. Patients were assigned 160 mg of aspi-
absolute benefits and risks of aspirin in the primary rin or placebo qd for 5 weeks, with the first dose
prevention trials were very small. Each year, fewer starting before surgery. Other forms of prophylaxis
than 1 person in every 1,000 could expect to avoid an were allowed, and either heparin or low-molecular-
occlusive vascular event by taking aspirin, whereas a weight heparin was used in about 40% of the
patients. Among the 13,356 patients with hip frac-
comparably small number could expect to experience a
ture, aspirin produced a 36% reduction in symptom-
major extracranial bleed. The relative size of these
atic deep vein thrombosis or pulmonary embolism
opposing effects is too imprecisely known in low-risk
(PE) [absolute risk reduction 0.9%; p ⫽ 0.0003]. A
people to predict the net public health consequences of similar relative risk reduction in patients who were
widespread aspirin use in healthy people. The ATT assigned aspirin was observed in patients who also
Collaboration is currently analyzing individual partici- received heparin.
pant data from the six primary prevention trials, and This important study,138 therefore, clearly shows
these new analyses will help to clarify the benefits and that aspirin reduces the incidence of fatal PE and
risks of aspirin in particular groups of individuals. Until symptomatic nonfatal deep vein thrombosis or PE in
the benefits of aspirin can be defined more precisely, patients with hip fracture. The results of the Pulmo-
however, the possibility of a benefit does not seem to nary Embolism Prevention trial are consistent with
justify the probability of a hazard. This emphasizes the the meta-analysis performed by the Antiplatelet Tri-
need for trials of aspirin for primary prevention among alists’ Collaboration,139 and supersede the findings in
specific groups at increased risk of vascular disease, most of the previous trials.140 –142 However, in three
such as people ⬎ 70 years of age and people with randomized studies in major orthopedic surgery
diabetes but no vascular disease. comparing aspirin with either warfarin143 or a low-
molecular-weight heparin,144,145 the incidence of ve-
2.5.1 Atrial Fibrillation: Moderate-dose warfarin nous thrombosis was significantly higher in the aspi-
alone (international normalized ratio [INR], 2.0 to rin group in all three.
3.0) is very effective in reducing the risk of stroke in
patients with nonvalvular atrial fibrillation.129 –131 2.5.3 Placental Insufficiency: The pathogenesis of
The effectiveness of aspirin in doses between 75 and preeclampsia and fetal growth retardation is related
325 mg has been compared with warfarin and pla- to reduced placental blood flow, which is believed to
cebo in three randomized trials of patients with non- be caused by constriction and/or thrombosis of small
valvular atrial fibrillation.130,132,133 In one study,130 placental arteries.146 The initial reports that low-dose
aspirin was significantly more effective than placebo, aspirin therapy reduces the risk of severe low birth
whereas in the other two132,133 there was a nonsig- weight among newborns,147 and the risk of cesarean
nificant trend in favor of aspirin. Pooled analysis of section in mothers with pregnancy-induced hyper-
the three studies shows a relative risk reduction in tension,146 led to the widespread use of prophylactic
favor of aspirin over placebo of about 25% (range, 14 aspirin to prevent preeclampsia. Subsequently, sev-
to 44%). Aspirin was significantly less effective than eral larger trials reported no beneficial effects of
warfarin in two studies on an intention-to-treat anal- aspirin.148 –154
ysis,132,133 and in the third study on an efficacy A systematic review155 of data from 39 trials in
analysis.130 On pooled analysis, warfarin was signifi- ⬎ 30,000 women showed that antiplatelet therapy
cantly more effective than aspirin, with a 47% rela- (mostly aspirin, 60 mg/d) is associated with a 15%
tive risk reduction (range, 28 to 61%; p ⬍ 0.01).134 decrease in the risk of preeclampsia. This effect was
Moreover, adjusted-dose warfarin therapy (INR, 2.0 consistent, regardless of risk status (moderate or
to 3.0) was more effective than fixed low-dose war- high), dose of aspirin, or gestation at trial entry.
farin therapy (INR, 1.2 to 1.5) and aspirin, 325 mg/d, There was some evidence that there may be greater
in high-risk patients with atrial fibrillation.135 Thus, benefits for women given ⬎ 75 mg of aspirin, al-
aspirin appears to be effective in preventing stroke in though the numbers of women in the subgroup were
patients with atrial fibrillation but is substantially less small and so a potential for random error. There was
effective than warfarin.136,137 also an 8% reduction in the risk of preterm birth and

a 14% reduction in the risk of fetal or neonatal death mately 1/100th the absolute number of major vascu-
for women allocated antiplatelet therapy.155 Remain- lar events avoided by aspirin therapy.7
ing questions are whether particular subgroups of The overall risk of major extracranial and intracra-
high-risk women might have greater benefit and nial hemorrhage associated with antiplatelet drugs is
whether earlier treatment (ie, before 12 weeks) or difficult to assess in individual trials because their
aspirin doses of ⬎ 75 mg would have additional incidence is low (ie, ⬍ 1%/yr), making detection of
benefits without an increase in adverse effects.155 even a 50 to 60% relative increase in risk unrealistic
The potential involvement of extra platelet sources of in most trials of a few thousand patients.
vasoactive eicosanoids expressing COX-2 in response Aspirin-induced GI toxicity, as detected in ran-
to a local growth-promoting milieu might contribute, domized clinical trials, appears to be dose related in
at least in part, to the limited efficacy of low-dose the range of 30 to 1,300 mg/d.156 This, along with
aspirin therapy in this setting. studies of the relationship of efficacy to dose, is
based largely on indirect comparisons of different
2.6 Adverse Effects of Aspirin trials and on a limited number of randomized, direct
comparisons of different aspirin doses, as reviewed
Aspirin does not cause a generalized bleeding above. Such a dose-response relationship is thought
abnormality unless it is given to patients with an to reflect at least two COX-1-dependent compo-
underlying hemostatic defect, such as hemophilia, nents, dose-dependent inhibition of COX-1 in the GI
uremia, or that induced by anticoagulant therapy. mucosa and dose-independent (within the range of
Aspirin-induced impairment of primary hemostasis examined doses) inhibition of COX-1 in platelets.6
cannot be separated from its antithrombotic effect Thus, it is not surprising that the antithrombotic
and is similar at all doses ⱖ 75 mg/d.6 effect of aspirin can be dissociated, at least in part,
The balance between preventing vascular occlu- from its most common side effect. However, even
sion and causing excess bleeding with aspirin de- when administered at low doses, aspirin can cause
pends critically on the absolute thrombotic vs hem- serious GI bleeding, as reported in studies using 30
orrhagic risk of the patient. Thus, in individuals at to 50 mg/d.36,42 Because of the underlying preva-
low risk for vascular occlusion (eg, ⱕ 1%/yr), a very lence of gastric mucosal erosions related to concur-
small absolute benefit is offset by exposure of a large rent use of other NSAIDs and/or Helicobacter pylori
number of healthy subjects to undue bleeding com- infection in the general population, it should be
plications. In contrast, in patients at high risk of expected that any antiplatelet dose of aspirin will
cardiovascular or cerebrovascular complications (eg, cause more bleeding from preexisting lesions than a
⬎ 3%/yr), the substantial absolute benefit of aspirin placebo. Consistent with this mechanistic interpre-
prophylaxis clearly outweighs the harm (Table 5). tation, the relative risk of hospitalization due to
For example, the absolute excess of major bleeds (ie, upper-GI bleeding and/or perforation associated
those requiring transfusion) in acute MI is approxi- with low-dose aspirin therapy (mostly, 100 to 300
mg/d) is comparable to that of other antiplatelet
agents and anticoagulants (ie, 2.3 [95% CI, 1.7 to
3.2], 2.0 [95% CI, 1.4 to 2.7], and 2.2 [95% CI, 1.4 to
Table 5—Benefit and Harm of Antiplatelet Prophylaxis 3.4], respectively) in a large population-based obser-
With Aspirin in Different Settings
(Section 2.6) vational study.157
In the overview of the ATT Collaboration,7 infor-
Clinical Setting Benefits* Harm† mation was available on 787 major extracranial hem-
Men at low to high cardiovascular risk 1–2 1–2 orrhages in 60 trials recording at least one such
Essential hypertension 1–2 1–2 hemorrhage. These were generally defined as hem-
Chronic stable angina 10 1–2 orrhages that were fatal or required transfusion;
Prior MI 20 1–2
among them, 159 (20%) caused death. Overall, the
Unstable angina 50 1–2
proportional increase in risk of a major extracranial
*No. of patients in whom a major vascular event is avoided per
bleed with antiplatelet therapy was about one half
1,000/yr. Benefits are calculated from randomized trial data re-
viewed in this article and depicted in Figure 3. (odds ratio [OR], 1.6; 95% CI, 1.4 to 1.8), with no
†No. of patients in whom a major GI bleeding event is caused per significant difference between the proportional in-
1,000/yr. Excess of upper-GI bleedings is estimated from a back- creases observed in each of the five high-risk cate-
ground rate of 1 event per 1,000/yr in the general population of gories of patients. After allowing for noncompliance
nonusers157 and an RR of 2.0 to 3.0 associated with aspirin
in the trials, they are compatible with the 2- to
prophylaxis.157–159 Such an estimate assumes comparability of other
risk factors for upper-GI bleeding, such as age and concomitant use 2.5-fold excess observed in case-control studies.
of NSAIDs, and may actually underestimate the absolute risk in an A case-control study with hospital and community
elderly population exposed to “primary” prevention. controls has examined the risks of hospitalization for

bleeding peptic ulcer associated with three different current guidelines that recommend clopidogrel for
regimens of aspirin prophylaxis.158 ORs were raised patients who have major GI contraindications to
for all doses of aspirin taken: 75 mg, OR 2.3 (95% CI, aspirin, principally recent significant bleeding from a
1.2 to 4.4); 150 mg, OR 3.2 (95% CI, 1.7 to 6.5); peptic ulcer or gastritis.169,170 Both studies enrolled
and 300 mg, OR 3.9 (95% CI, 2.5 to 6.3). Addi- patients with ulcer bleeding after the use of low-dose
tional epidemiologic studies have found a dose- aspirin. In the study of Chan et al,167 after healing of
response relationship between aspirin prescription ulcers and eradication of H pylori, if present, 320
and upper-GI complications, as reviewed by Garcı́a patients were randomly assigned to receive either
Rodrı́guez et al.159 It has been calculated that ap- clopidogrel, 75 mg/d, or aspirin, 80 mg/d, plus 20 mg
proximately 900 of the 10,000 episodes of ulcer bid of esomeprazole for 12 months. The cumulative
bleeding occurring in people ⬎ 60 years of age each incidence of recurrent bleeding was 8.6% (95% CI,
year in England and Wales could be associated with, 4.1 to 13.1%) among patients who received clopi-
and ascribed to, prophylactic aspirin use.158 A gen- dogrel and 0.7% (95% CI, 0 to 2.0%) among those
eral change to lower doses of aspirin (75 mg) would who received aspirin plus esomeprazole (p ⫽ 0.001).167
not eliminate risks but would reduce risk by about In the study of Lai et al,168 170 patients with prior ulcer
40% compared with 300-mg doses and by 30% bleeding were randomly assigned to treatment with
compared with 150-mg doses if the assumptions clopidogrel, 75 mg/d, or aspirin, 100 mg/d, and esome-
from indirect comparisons are correct.158 Given that prazole, 20 mg/d, for 1 year. The cumulative incidence
the mortality rate among patients who are hospital- of recurrent ulcer complications was 13.6% and 0%,
ized for NSAID-induced upper-GI bleeding is about respectively (95% CI for the difference, 6.3 to 20.9%;
5 to 10%,160,161 such a strategy could save a signifi- p ⫽ 0.0019).168 The consistent findings of two indepen-
cant number of lives. dent studies suggest that the combination of esomepra-
The widely held belief that enteric-coated and zole and low-dose aspirin is superior to clopidogrel in
buffered varieties of aspirin are less likely to occasion preventing recurrent ulcer bleeding in patients with a
major upper-GI bleeding than plain tablets was history of aspirin-related ulcer bleeding.
tested in data from a multicenter case-control Substantially less information is available about
study.162 The relative risks of upper-GI bleeding for the risk of intracranial hemorrhage associated with
plain, enteric-coated, and buffered aspirin at average aspirin use. In the Nurses’ Health Study171 cohort of
daily doses of ⱕ 325 mg were 2.6, 2.7, and 3.1, approximately 79,000 women 34 to 59 years of age,
respectively. At doses ⬎ 325 mg, the relative risks infrequent use of aspirin (1 to 6 tablets per week)
were 5.8 for plain and 7.0 for buffered aspirin; there was associated with reduced risk of ischemic stroke,
were insufficient data to evaluate enteric-coated whereas high frequency of use (ⱖ 15 aspirin tablets
aspirin at this dose level.162 Similar conclusions were per week) was associated with increased risk of
reached by a case-control study using data from the subarachnoid hemorrhage, particularly among older
UK General Practice Research Database.163 Thus, or hypertensive women. In the overview of the ATT
physicians who recommend aspirin in an enteric- Collaboration,7 the overall absolute excess of intra-
coated or buffered form should not assume that cranial hemorrhage due to aspirin therapy was ⬍ 1
these formulations are less likely to cause GI tract per 1,000 patients per year in high-risk trials, with
bleeding than plain aspirin. somewhat higher risks in patients with cerebrovas-
Suppressing acid secretion is thought to reduce cular disease.
the risk of ulcers associated with regular use of Low-dose aspirin therapy has not been reported to
NSAIDs. In patients who required continuous treat- affect renal function or BP control,172 consistent with
ment with NSAIDs and who had ulcers or ⬎ 10 its lack of effect on renal prostaglandins173 that derive
erosions in either the stomach or duodenum, ome- primarily from constitutively expressed COX-2 in the
prazole healed and prevented ulcers more effectively human kidney.83 Moreover, aspirin, 75 mg/d, did not
than did ranitidine.164 In these patients, maintenance affect BP or the need for antihypertensive therapy in
therapy with omeprazole was associated with a lower intensively treated hypertensive patients.122 The sug-
rate of relapse and was better tolerated than miso- gestion that the use of aspirin and other antiplatelet
prostol.165 In high-risk patients (history of previous agents is associated with reduced benefit from ena-
ulcer bleeding) taking low-dose aspirin for 6 months, lapril in patients with left ventricular systolic dys-
omeprazole and H pylori eradication were associated function174 is not supported by the results of a large
with similar rates of recurrent bleeding (0.9% vs metaanalysis of MI trials.175 Similarly, no negative
1.9%),166 although clinically important differences interaction occurs between angiotensin-converting
between the two preventive strategies could not be enzyme (ACE) inhibition and the cardiovascular
excluded owing to the small sample size (n ⫽ 250). benefits of low-dose aspirin in intensively treated
Two relatively small studies167,168 have challenged hypertensive patients.176 The ACE Inhibitors Col-

laborative Group177 has performed a systematic the antithrombotic effect of aspirin does not appear
overview of data for 22,060 patients from six long- to be dose related over a wide range of daily doses
term randomized trials of ACE inhibitors to assess (30 to 1,300 mg), an observation consistent with
whether aspirin altered the effects of ACE inhibitor saturability of platelet COX inhibition at very low
therapy on major clinical outcomes. Even though doses. In contrast, GI toxicity of the drug does
results from these analyses cannot rule out the appear to be dose related, consistent with dose- and
possibility of some sort of interaction, they show dosing interval-dependent inhibition of COX activity
unequivocally that even if aspirin is given, the addi- in the nucleated lining cells of the GI mucosa. Thus,
tion of ACE inhibitor therapy produced substantial aspirin once daily should be considered in all clinical
additional benefit in all major vascular outcomes. conditions in which antiplatelet prophylaxis has a
Therefore, in the absence of clear contraindications, favorable benefit/risk profile. Because of GI toxicity
concomitant use of aspirin and ACE inhibitors and its potential impact on compliance, physicians
should be considered in all patients at high risk of are encouraged to use the lowest dose of aspirin
major vascular events.177 shown effective in each clinical setting (Table 1).
Thus, in summary, inhibition of TXA2-dependent
platelet function by aspirin is effective for the pre-
2.7 Reversible COX Inhibitors
vention of thrombosis, but is also associated with
excess bleeding. Assessing the net effect requires an In the absence of definitive randomized studies,
estimation of the absolute thrombotic vs hemor- traditional NSAIDs have long been thought to pose
rhagic risk of the individual patient. In individuals at no cardiovascular hazard or to be somewhat cardio-
very low risk for vascular occlusion, a very small protective. Because of their reversible mechanism of
absolute benefit may be offset by exposure of very action in inhibiting platelet COX-1 and of their short
large numbers of healthy subjects to undue bleeding half-lives, most traditional NSAIDs inhibit TXA2-
complications. As the risk of experiencing a major dependent platelet activation only transiently and
vascular event increases, so does the absolute benefit incompletely in the vast majority of users.178 A
of antiplatelet prophylaxis with aspirin, as shown in notable exception is provided by naproxen, which
Figure 3, for a number of clinical settings in which when administered regularly at 500 mg bid, has been
the efficacy of the drug has been tested in random- shown to inhibit TXA2 biosynthesis in vivo to the
ized clinical trials. Based on the results of such trials, same extent as low-dose aspirin,179 consistent with its
60
50 Unstable Angina
Subjects 40
in whom a
vascular event 30
is prevented by
aspirin
20
per 1,000 treated Survivors of MI
/yr
10 Stable Angina
Healthy Subjects
0
0 5 10 15 20
Annual risk of a vascular event on placebo
Figure 3. The absolute risk of vascular complications is the major determinant of the absolute benefit
of antiplatelet prophylaxis. Data are plotted from placebo-controlled aspirin trials in different clinical
settings. For each category of patients, the abscissa denotes the absolute risk of experiencing a major
vascular event as recorded in the placebo arm of the trial(s). The absolute benefit of antiplatelet
treatment is reported on the ordinate as the number of subjects in whom an important vascular event
(nonfatal MI, nonfatal stroke, or vascular death) is actually prevented by treating 1,000 subjects with
aspirin for 1 year.

relative COX-1 selectivity and longer half-life than Triflusal, a salicylic acid derivative, reversibly in-
other commonly used NSAIDs. hibits platelet COX activity after conversion to a
The only reversible COX inhibitors that have been long-lived metabolite, 2-hydroxy-4-trifluoromethyl-
tested in randomized clinical trials for their antithrom- benzoic acid.189 Although the half-life of the parent
botic efficacy are sulfinpyrazone, indobufen, flurbipro- compound is only about 30 min, that of the deacety-
fen, and triflusal. Sulfinpyrazone is a uricosuric agent lated metabolite approximates 2 days. Although tri-
structurally related to the antiinflammatory agent phe- flusal is claimed to have negligible effects on vascular
nylbutazone. When used at the highest approved PGI2 production, this is likely to reflect the exper-
dosage of 200 mg qid, the drug inhibits platelet COX imental conditions used for the assessment of
activity by approximately 60%, after conversion from PGI2 production ex vivo. The limited sample size
an inactive sulfoxide to an active sulfide metabo- of head-to-head comparisons of triflusal vs aspirin
lite.180 The conflicting or negative results obtained in in patients randomized within 24 h of acute MI190
randomized clinical trials of sulfinpyrazone in pa- and in patients with cerebrovascular disease191
tients with MI or unstable angina7 (reviewed in the precludes unequivocal interpretation of the similar
“Valvular and Structural Heart Disease” chapter) are rates of major vascular events in the two treatment
not surprising in light of the drug being a weak COX groups. None of these reversible COX inhibitors
inhibitor with no other established antiplatelet are approved as an antiplatelet drug in the United
mechanism of action. States, and it is unclear under which circum-
In contrast, indobufen is a very potent inhibitor of stances they are prescribed instead of aspirin in
platelet COX-1 activity and has comparable bio- other countries.
chemical, functional, and clinical effects to those of a
standard dose of aspirin. Thus, at therapeutic plasma
2.8 Coxibs and Cardiovascular Disease
levels achieved after oral dosing of 200 mg bid,
indobufen inhibits serum TXB2 by ⬎ 95% through- Coxibs were developed in an attempt to prevent
out the dosing interval181 and reduces urinary TX the adverse GI effects of nonselective NSAIDs (by
metabolite excretion to an extent quite comparable avoiding inhibition of COX-1) while maintaining
to aspirin.182 The finding that indobufen is as effec- equivalent antiinflammatory efficacy (by inhibiting
tive as aspirin in preventing coronary graft occlusion COX-2).83 Several large randomized trials192–194 have
in two randomized trials183,184 is mechanistically demonstrated that coxibs are associated with lower
consistent with the concept of platelet COX-1 inhi- risk of serious GI events than nonselective NSAIDs,
bition largely accounting for the antithrombotic ef- but the Vioxx GI Outcomes Research Study193
fect of aspirin, as discussed above. Indobufen also among approximately 8,000 patients with rheuma-
has been investigated in a small placebo-controlled toid arthritis showed that those allocated to rofe-
study of patients with heart disease at increased coxib, 50 mg/d, experienced a higher risk of vascular
embolic risk185 and compared with warfarin186 and events than those allocated to naproxen 500 mg bid.
ticlopidine187 in patients with nonrheumatic atrial This excess was almost entirely accounted for by a
fibrillation and patients with recent reversible cere- difference in the incidence of MI (20 in 2,699
bral ischemia, respectively. However, none of these person-years of follow-up among rofecoxib-allocated
studies in ⬎ 4,000 patients clearly established an patients, vs 4 in 2,699 person-years among naproxen-
advantage of indobufen vs standard treatments, al- allocated patients). There were no significant differ-
though the 95% CIs for these comparisons are wide. ences in stroke (11 rofecoxib vs 9 naproxen) or
Indobufen has been reported to suppress in vivo vascular deaths (7 rofecoxib vs 7 naproxen).193 Three
TXA2 biosynthesis more effectively than low-dose placebo-controlled trials have now revealed a two-
aspirin in patients with unstable angina, an effect fold- to threefold-increased risk of vascular events in
possibly related to inhibition of monocyte COX-2 by approximately 6,000 patients treated short term (10
therapeutic plasma levels of indobufen.14 The clini- days) with valdecoxib195 or long term (up to 3 years)
cal relevance of these findings remains to be estab- with celecoxib196 or rofecoxib197 both with and with-
lished. out concomitant aspirin treatment. These recent
Flurbiprofen has been evaluated in a single placebo- findings are consistent with a mechanism-based car-
controlled, randomized trial of 461 patients with diovascular hazard for the class198 and have led to the
acute MI.188 The 6-month reinfarction rate was withdrawal of rofecoxib and valdecoxib from the
significantly lower in the flurbiprofen group (3%) market.
than in the placebo group (10.5%), with an extremely A metaanalysis of tabular data from 138 random-
low mortality rate (1.1%) in both groups. The small ized trials of five different coxibs in approximately
sample size of the study limits interpretation of these 145,000 patients has revealed that in placebo com-
findings. parisons, allocation to a coxib was associated with a

42% increased incidence of vascular events with no ridamole concentrations required to produce these
statistically significant heterogeneity among the dif- effects far exceed the low micromolar plasma levels
ferent coxibs.20 This excess risk of vascular events achieved after oral administration of conventional
was derived primarily from a twofold-increased risk doses (100 to 400 mg/d).200 Dipyridamole also dif-
of MI. Overall, there was no significant difference in ferentially inhibits the expression of critical inflam-
the incidence of vascular events between a coxib and matory genes in platelet-leukocyte aggregates.201
any traditional NSAID, but there was evidence of a The absorption of dipyridamole from conventional
significant difference between naproxen and the formulations is quite variable and may result in low
other traditional NSAIDs.20 Given the nonlinear systemic bioavailability of the drug. A modified-
relationship between inhibition of platelet COX-1 release formulation of dipyridamole with improved
activity and inhibition of platelet activation in vivo bioavailability has been developed in association with
(Fig 2),31 it is perhaps not surprising that the cardio- low-dose aspirin.202 Dipyridamole is eliminated pri-
vascular safety profile of coxibs and some non- marily by biliary excretion as a glucuronide conjugate
naproxen NSAIDs (primarily diclofenac and ibupro- and is subject to enterohepatic recirculation. A ter-
fen) appears similar because these drugs fail to minal half-life of 10 h has been reported. This is
inhibit platelet activation adequately irrespective of consistent with the twice-daily regimen used in
their COX-2 selectivity. The results of the Multina- recent clinical studies.
tional Etoricoxib and Diclofenac Arthritis Long- Although the clinical efficacy of dipyridamole,
Term study,199 comparing long-term treatment with alone or in combination with aspirin, has been
etoricoxib and diclofenac in approximately 35,000 questioned on the basis of earlier randomized tri-
arthritis patients, are consistent with this conclusion. als,2,203 the whole issue has been reopened by the
Whether the variable level and duration of COX-1 reformulation of the drug to improve bioavailability
inhibition by different NSAIDs modulate the cardio- and the results of the ESPS-2 and European Stroke
vascular consequences of COX-2 inhibition presently Prevention Reversible Ischemia Trial (ESPRIT)
is unknown, given the limited utilization of NSAIDs studies.36,204 In ESPS-2, the new preparation of
other than ibuprofen, diclofenac, and naproxen in dipyridamole was evaluated in 6,602 patients with
coxib trials. Thus, coxibs and some traditional NSAIDs prior stroke or TIA.36 This study showed that the
moderately increase the risk of vascular events, addition of modified-release dipyridamole 200 mg
particularly MI, but there remains considerable un- bid to aspirin 25 mg bid was associated with a 22%
certainty about the magnitude of this hazard for relative risk reduction of major vascular events com-
particular drug regimens and patients subgroups. A pared with aspirin alone. Headache was the most
metaanalysis of individual participant data from ran- common adverse effect of dipyridamole. Bleeding at
domized coxib trials is currently being conducted by any site was almost doubled in the two aspirin arms
the Coxib Trialists’ Collaboration in order to address but was surprisingly indistinguishable from placebo
some of the open questions related to the influence in the dipyridamole-treated patients.36 In a post hoc
of dose, duration, and baseline characteristics, in- analysis of cardiac events in patients with CHD or
cluding the concomitant use of low-dose aspirin, on MI at entry, dipyridamole did not result in a higher
this cardiotoxicity. number of fatal and nonfatal cardiac events.204
More recently, the ESPRIT Study Group205 has
performed a randomized trial in which they assigned
3.0 Dipyridamole 2,739 patients within 6 months of a TIA or minor
stroke of presumed arterial origin to aspirin (30 to
Dipyridamole is a pyrimidopyrimidine derivative 325 mg/d) with or without dipyridamole (200 mg
with vasodilator and antiplatelet properties. The bid). The primary outcome (a composite of major
mechanism of action of dipyridamole as an antiplate- vascular events or major bleeding complications) was
let agent has been a subject of controversy.200 Both significantly reduced by the combined treatment vs
inhibition of cyclic nucleotide phosphodiesterase aspirin alone by 20%. Patients receiving aspirin and
(the enzyme that degrades cyclic adenosine mono- dipyridamole discontinued trial medication almost
phosphate 关AMP兴 to 5(1)-AMP, resulting in the three times more often than those receiving aspirin
intraplatelet accumulation of cyclic AMP, a platelet alone, mainly because of headache.205 Addition of
inhibitor) and blockade of the uptake of adenosine the ESPRIT data to the metaanalysis of previous
(which acts at A2 receptors for adenosine to stimu- trials resulted in an overall risk ratio of 0.82 (95% CI,
late platelet adenylyl cyclase and thus increase cyclic 0.74 to 0.91) for the composite of vascular death,
AMP) have been suggested. Moreover, direct stim- stroke, or MI. However, based on the most recent
ulation of PGI2 synthesis and protection against its Cochrane review,203 the additional benefit of the
degradation have been reported, although the dipy- combination over aspirin alone is not detectable in

patients with other types of vascular disease. daily dosing with ticlopidine or clopidogrel and
Whether this apparent discrepancy reflects a differ- with slow recovery of platelet function after drug
ent prevalence of dipyridamole-sensitive mecha- withdrawal.206
nisms of disease or, perhaps more likely, the differ-
ent types of formulation and daily dosage of the drug
4.1 Ticlopidine
remains to be established. The fixed combination of
modified-release dipyridamole and low-dose aspirin Up to 90% of a single oral dose of ticlopidine is
has been approved for stroke prevention by the FDA rapidly absorbed in humans.206 Peak plasma concen-
and other regulatory authorities. trations occur 1 to 3 h after a single oral dose of 250
mg. Plasma levels of ticlopidine increase by approx-
imately threefold on repeated twice-daily dosing
4.0 Thienopyridines over 2 to 3 weeks because of drug accumulation.
Greater than 98% of ticlopidine is reversibly bound
Ticlopidine and clopidogrel are structurally re- to plasma proteins, primarily albumin. Ticlopidine is
lated thienopyridines with platelet-inhibitory prop- metabolized rapidly and extensively. A total of 13
erties. Both drugs selectively inhibit ADP-induced metabolites have been identified in humans. Of
platelet aggregation with no direct effects on arachi- these, only the 2-keto derivative of ticlopidine is
donic acid metabolism.206 Although ticlopidine and more potent than the parent compound in inhibiting
clopidogrel also can inhibit platelet aggregation in- ADP-induced platelet aggregation.206
duced by collagen and thrombin, these inhibitory The apparent elimination half-life of ticlopidine is
effects are abolished by increasing the agonist con- 24 to 36 h after a single oral dose and up to 96 h after
centration and, therefore, are likely to reflect block- 14 days of repeated dosing.206 The standard regimen
ade of ADP-mediated amplification of the platelet of ticlopidine is 250 mg bid, although it is unclear
response to other agonists. how a twice-daily regimen is related to the pharma-
Neither ticlopidine nor clopidogrel affect ADP- cokinetic and pharmacodynamic features noted
induced platelet aggregation when added in vitro, up above. A delayed antithrombotic effect was noted in
to 500 ␮mol/L, thus suggesting that in vivo hepatic at least one clinical trial of ticlopidine in patients
transformation to an active metabolite(s) is necessary with unstable angina with no apparent protection
for their antiplatelet effects. In the liver, clopidogrel during the first 2 weeks of drug administration.213
is metabolized into 2-oxo-clopidogrel through a cy- Therefore, ticlopidine is not useful when a rapid
tochrome P450-dependent pathway. This intermedi- antiplatelet effect is required.
ate metabolite is then hydrolyzed and generates the Ticlopidine as a single agent has been evaluated in
highly labile active metabolite,207 which reacts as a patients with stroke,214 transient cerebral isch-
thiol reagent with the ADP receptors on platelets emia,215 unstable angina,213 MI,216 intermittent clau-
when they pass through the liver.208 The active dication,217–219 and aortocoronary bypass surgery.220
metabolite belongs to a family of eight stereoiso- Ticlopidine was significantly (but marginally in abso-
mers, only one of which (bearing 7S, 3Z, and 4S or lute terms) more effective than aspirin in reducing
4R configuration) retains biological activity.208 stroke in patients with transient cerebral ischemia or
Experimental evidence suggests that clopidogrel minor stroke215 (although there was no statistically
and, probably, ticlopidine induce irreversible alter- significant difference in the combined outcome of
ations of the platelet receptor P2Y12 mediating stroke, MI, or death)7; was as effective as aspirin in
inhibition of stimulated adenylyl cyclase activity by the treatment of patients with a recent MI216; was
ADP.209,210 The active metabolite of clopidogrel more effective than placebo in reducing the risk of
couples through a disulfide bridge to the P2Y12 the combined outcome of stroke, MI, or vascular
receptor, presumably to the cysteine residue in the death in patients with thromboembolic stroke214; was
first extracellular loop; this results in oligomers dis- more effective than conventional antianginal therapy
sociating into dimeric receptors that are partitioned in reducing vascular death or MI in patients with
out of lipid rafts, thereby losing the ability to bind unstable angina213; was more effective than placebo
their endogenous ligand.211 Interestingly, mutations in reducing acute occlusion of coronary bypass
in the P2Y12 gene are associated with a congenital grafts220; and was more effective than controls in
bleeding disorder and abnormality in the platelet improving walking distance218 and reducing vascular
response to ADP, resembling that induced by complications in patients with peripheral vascular
thienopyridines.212 Permanent modification of a disease.217–219 The association of ticlopidine therapy
platelet ADP receptor by thienopyridines is con- with hypercholesterolemia and neutropenia (for
sistent with time-dependent cumulative inhibition which the reported rate of occurrence is 2.4% for
of ADP-induced platelet aggregation on repeated neutrophils ⬍ 1.2 ⫻ 109/L and 0.8% for neutrophils

⬍ 0.45 ⫻ 109/L) and its comparative expense have the carboxylic acid derivative, SR 26334. Based on
reduced enthusiasm for this therapy as an alternative measurements of circulating levels of SR 26334, it has
to aspirin in most situations.221 Ticlopidine also has been inferred that clopidogrel is rapidly absorbed and
been associated with thrombocytopenia,221 aplastic extensively metabolized.230 The plasma elimination
anemia,222 and thrombotic thrombocytopenic pur- half-life of SR 26334 is approximately 8 h. As noted
pura (TTP).223 Ticlopidine has been approved for above, clopidogrel, inactive in vitro, is metabolically
clinical use in patients with cerebral ischemia when transformed by the liver into a short-lived active plate-
aspirin has failed, cannot be tolerated, or is contra- let inhibitor. However, the interindividual variability in
indicated, although this limitation does not apply to this metabolic activation is still being assessed, and to
all countries where the drug is registered. our knowledge, there are no published data on whether
Additive effects of ticlopidine and aspirin have liver impairment decreases the ability of clopidogrel to
been described in rats, in inhibition of ADP-induced inhibit platelet function. A pharmacokinetic/pharmacody-
platelet aggregation ex vivo, tail bleeding time pro- namic study232 of a 600-mg loading dose of clopi-
longation, and protection from thrombosis in exper- dogrel in healthy subjects revealed linear correla-
imental models of platelet-dependent vascular occlu- tions between the maximal antiplatelet effect and
sion.224 Additive antiplatelet effects of aspirin (40 peak plasma concentrations of unchanged clopi-
mg) and ticlopidine (250 mg) have been reported in dogrel, of the carboxyl metabolite, and of the thiol
healthy volunteers.225 Two studies226,227 have dem- metabolite as well as linear correlations between
onstrated the superiority of ticlopidine with aspirin peak plasma concentration values of clopidogrel and
compared to aspirin alone or aspirin plus warfarin in its metabolites. These results have been interpreted
preventing thrombotic complications after coronary to suggest that the pharmacodynamic response vari-
artery stent placement. Ticlopidine has been rou- ability is predominantly caused by individual differ-
tinely used in combination with aspirin in patients ences in clopidogrel absorption.232 Because the
receiving coronary artery stents, but the better safety cytochrome P450 isozymes CYP3A4 and 3A5 metab-
profile of clopidogrel has resulted in the replacement olize clopidogrel faster than other human P450
of clopidogrel for ticlopidine as the standard anti- isozymes and are the most abundant P450s in human
platelet regimen after stent deployment.228 The risk liver, they are predicted to be predominantly respon-
of TTP associated with ticlopidine use has been sible for the activation of clopidogrel in vivo.233
estimated as 0.02% in patients receiving the drug When clopidogrel and atorvastatin, a CYP3A4 sub-
after stent placement.229 This risk compares with an strate, are present at equimolar concentrations in
incidence of 0.0004% in the general population. The vitro, clopidogrel metabolism is inhibited by
mortality rate for this rare complication exceeds ⬎ 90%.233 Variable metabolic activity of CYP3A4
20%.229 The place of ticlopidine in the current may contribute to the interindividual variability in
therapeutic armamentarium is uncertain: (1) the the platelet inhibitory effect of clopidogrel.234 Thus,
drug is not uniformly cheaper than clopidogrel in Angiolillo et al235 recently characterized the influ-
different countries; (2) in contrast to clopidogrel, ence of CYP3A4 genotype on interpatient variability
ticlopidine has no approved indication for the long- in clopidogrel responsiveness. An intronic single
term management of post-MI patients; (3) ticlopi- nucleotide polymorphism in the CYP3A4 gene,
dine has a higher bone-marrow toxicity than clopi- IVS10 ⫹ 12G ⬎ A (also called CYP3A4*1G) modified
dogrel; and (4) because of safety concerns, an platelet reactivity ex vivo as measured by GP-IIb/IIIa
adequate loading dose of ticlopidine, as required in receptor activation in response to clopidogrel in a group
the acute setting, is unlikely to be used. of patients with stable CHD receiving long-term anti-
platelet therapy. The findings were replicated in a
group of clopidogrel-naı̈ve patients undergoing elective
4.2 Clopidogrel
percutaneous coronary intervention (PCI) treated with
The pharmacokinetics of clopidogrel are somewhat a 300-mg loading dose of the drug.235
different from those of ticlopidine. Thus, after admin- Clopidogrel inhibited ADP-induced platelet ag-
istration of single oral doses (up to 200 mg) or repeated gregation in a dose-dependent fashion with an ap-
doses (up to 100 mg/d), unchanged clopidogrel was not parent ceiling effect (40% inhibition) at 400 mg after
detectable in peripheral venous plasma.230 Concentra- single oral doses in healthy volunteers. Inhibition of
tions of 1 to 2 ng/mL were measured in the plasma platelet aggregation was detectable 2 h after oral
of patients who received 150 mg/d of clopidogrel dosing of 400 mg and remained relatively stable up
(twice as much as the dose used in the Clopidogrel to 48 h.230 On repeated daily dosing of 50 to 100 mg
vs Aspirin in Patients at Risk of Ischemic Events of clopidogrel to healthy volunteers, ADP-induced
[CAPRIE] study231 and approved for clinical use) for platelet aggregation was inhibited from the second
16 days. The main systemic metabolite of clopidogrel is day of treatment (25 to 30% inhibition) and reached

a steady state (50 to 60% inhibition) after 4 to 7 days. cant clinical interaction between the two. However,
Such level of maximal inhibition was comparable to it should be emphasized that retrospective post hoc
that achieved with ticlopidine, 500 mg/d, although analyses have limitations that preclude definitive
the latter showed a slower onset of the antiplatelet conclusions. Moreover, the lack of information on
effect than the clopidogrel. No appreciable differ- statin daily doses used in trials notably restricts our
ences in the maximum inhibitory effects of 50, 75, ability to assess the dose dependence of potential
and 100 mg of clopidogrel were noted in this drug interactions.
study.230 As would be expected from these pharma- As with aspirin, both the mechanism(s) and the
cokinetic and pharmacodynamic features, a loading clinical relevance of clopidogrel resistance or nonre-
dose (eg, 300 mg) of clopidogrel results in a much sponsiveness251 remain to be established. Thus, no
more rapid onset of platelet inhibition than is test of platelet function can currently be recom-
achieved with the 75-mg dose.236 mended to assess the effects of clopidogrel in the
Several recent studies have examined the ade- individual patient, as there is no uniformly estab-
quacy of a 300-mg loading dose of clopidogrel in lished method for quantification of ex vivo platelet
patients scheduled for cardiac catheterization as reactivity after clopidogrel treatment and to what
potential candidates for PCI.237–240 After loading extent platelet activity is inhibited by the drug.251
with 600 mg of clopidogrel, the full antiplatelet Thus, the active metabolite of clopidogrel has a
effect of the drug was achieved after 2 h.237 More- pharmacodynamic pattern quite similar to that of
over, a loading dose of 600 mg resulted in higher aspirin in causing cumulative inhibition of platelet
plasma concentrations of the active metabolite, clo- function on repeated daily administration of low
pidogrel, and the carboxyl metabolite than did a doses. As in the case of aspirin, platelet function
loading dose of 300 mg.238 ADP-induced platelet returns to normal 7 days after the last dose of
aggregation also was significantly lower in patients clopidogrel. Both the cumulative nature of the in-
receiving 600 mg than in those receiving 300 hibitory effects and the slow rate of recovery of
mg.238 –240 The incremental antiplatelet effect of platelet function are consistent with the active moi-
900 mg over 600 mg of clopidogrel appears mar- eties of aspirin (acetylsalicylic acid) and clopidogrel
ginal,238,239 possibly because of limited drug ab- (active metabolite), causing a permanent defect in a
sorption.238 platelet protein that cannot be repaired during the
Clopidogrel treatment exhibited marked interindi- 24-h dosing interval and can only be replaced as a
vidual variability in inhibiting platelet function in function of platelet turnover. This consideration also
three different studies241–243 of patients undergoing justifies the once-daily regimen of both drugs, de-
elective PCI and stenting. A variable proportion of spite their short half-life in the human circulation. It
these patients were considered to be clopidogrel should be noted, however, that although aspirin
nonresponders or to have clopidogrel resistance currently is used at doses that represent a 2.5- to
based on ADP-induced platelet aggregation. Three 10-fold excess over the dose of 30 mg necessary
separate studies241,244,245 suggested that concurrent and sufficient to fully inactivate platelet COX-1
treatment with lipophilic statins that are substrates of activity on repeated daily dosing,25,62 clopidogrel is
CYP3A4 (eg, atorvastatin and simvastatin) may inter- used at doses causing only partial inactivation of
fere with the inhibitory effects of clopidogrel on P2Y12. Thus, the main determinants of the inter-
platelet function. In the study of Lau et al,244 individual variability in the antiplatelet effects of
atorvastatin, but not pravastatin, attenuated the an- the two drugs are substantially different (Table 6).
tiplatelet effect of clopidogrel in a dose-dependent Bleeding time measurements performed in the
manner. Because many drugs are metabolized by same multiple dose study230 described above showed
CYP3A4,246 it is likely that other drugs may modify a comparable prolongation (by 1.5- to 2.0-fold over
the systemic bioavailability of the active metabolite control) at 50 to 100 mg/d of clopidogrel or 500 mg/d
of clopidogrel and affect its clinical efficacy. More- ticlopidine.
over, variable metabolic activity of CYP3A4 may Clopidogrel has undergone a quite unusual clinical
contribute to the interindividual variability in the development, with very limited phase II studies and
platelet inhibitory effects of clopidogrel, as noted a single, very large phase III trial (ie, CAPRIE, to
above. Although ex vivo measurements of ADP-in- test its efficacy and safety at 75 mg/d vs aspirin at 325
duced platelet aggregation have suggested a pharma- mg/d).231 CAPRIE is unique among the studies that
cokinetic interaction between a CYP3A4-metabolized have directly compared antiplatelet agents against
statin and clopidogrel, post hoc analyses of placebo- aspirin in that it incorporated three groups of pa-
controlled studies 247,248 of clopidogrel, a single center tients, all of whom are recognized to be at an
cohort study,249 and data from a large multinational increased risk of recurrent ischemic events: those
registry250 have failed to detect a statistically signifi- who have experienced a recent stroke or recent MI

Table 6 —Main Determinants of the Interindividual The complementary mechanisms of action of clo-
Variability in the Antiplatelet Effects of Aspirin and pidogrel and low-dose aspirin has led to testing the
Clopidogrel (Section 4.2)
efficacy and safety of their combination in high-risk
Determinant Aspirin Clopidogrel clinical settings.253 The CURE trial253 randomly
Dependence on systemic bioavailability No Yes
assigned 12,562 patients with acute coronary syn-
Dependence on liver metabolism to No Yes dromes without ST-segment elevation who pre-
active moiety sented within 24 h after the onset of symptoms to
Recommended dose: minimum 2–3 1 receive clopidogrel (300-mg loading dose followed
effective dose for full by 75 mg qd) or placebo in addition to aspirin (75 to
pharmacodynamic effect
Relevance of pharmacodynamic Yes ?
325 mg/d) for 3 to 12 months. After a mean duration
interactions at the target site of treatment of 9 months, the primary outcome (a
Relevance of extraplatelet sources of Yes No composite of cardiovascular death, nonfatal MI, or
the platelet agonist stroke) occurred in 9.3% of the patients in the
clopidogrel group and 11.4% of the patients in the
placebo group (RR, 0.80; 95% CI, 0.72 to 0.90;
p ⬍ 0.001). The benefit of clopidogrel was apparent
and those presenting with symptomatic peripheral within the first 30 days after randomization and
arterial disease. Overall, CAPRIE showed a modest remained constant during the 12 months of the
difference in effectiveness between aspirin and clo- study. There were significantly more patients with
pidogrel that did not result in regulatory approval of major bleeding in the clopidogrel group than in the
a superiority claim. What is particularly interesting, placebo group (3.7% vs 2.7%; p ⫽ 0.001).
however, are the results obtained when the effects of The clinical benefit of dual antiplatelet therapy vs
aspirin and clopidogrel in the three groups are aspirin alone has been confirmed in patients under-
compared. Each comparison between clopidogrel going PCI,254 and in those presenting with an acute
and aspirin involved approximately 6,400 patients MI with ST-segment elevation within 12 h256 to 24
and, therefore, represents the largest head-to-head h257 after the onset of symptoms. In the Clopidogrel
comparison between aspirin and another antiplatelet and Metoprolol Myocardial Infarction Trial257
agent in that particular clinical setting, although the (COMMIT), addition of 75 mg/d of clopidogrel to
statistical power of such comparison is inadequate to 162 mg/d of aspirin reduced mortality and major
detect a modest difference between the two. This vascular events in the hospital by 9% (95% CI, 3 to
analysis shows that the majority of the difference in 14%), corresponding to nine fewer events per 1,000
effectiveness occurred in the patients who entered MI patients treated for about 2 weeks. Overall, when
the trial because of symptomatic peripheral arterial all transfused, fatal, or cerebral bleeds were consid-
disease. A formal test of heterogeneity of these three ered together, there was no significant excess risk
treatment effects was statistically significant (p ⫽ associated with the use of clopidogrel during the
0.042), suggesting that the true benefit of clopidogrel scheduled treatment period (0.58% clopidogrel plus
may not be identical across the three clinical settings. aspirin vs 0.55% aspirin alone; p ⫽ 0.59), nor was
Both clopidogrel and medium-dose aspirin ther- there any excess of major bleeds in patients ⬎ 70
apy were well tolerated in the CAPRIE study.231 The years of age or in those given fibrinolytic therapy
incidence of early permanent discontinuation of the before randomization.257 Clopidogrel was, however,
study drug due to adverse events was practically associated with a small, but significant, excess of 4.7
identical in the two treatment groups (12%). Simi- (95% CI, 1.4 to 8.0) reported minor bleeds per 1,000
larly, the overall incidence of hemorrhagic events patients treated. Taking major and minor bleeds
was identical in the aspirin and clopidogrel groups together, there was no apparent trend with respect
(9.3%). The frequency of severe rash and severe to age in the excess risk.257 Factors that may have
diarrhea was higher with clopidogrel than with aspi- contributed to the remarkable safety of dual anti-
rin, whereas GI discomfort and hemorrhage were platelet therapy in the COMMIT trial include the
more frequent with aspirin than with clopidogrel. No lack of a loading dose of clopidogrel, the uniform use
excess neutropenia was found in the clopidogrel of 162 mg of aspirin, and the short duration of
group, and thrombocytopenia was identical in the treatment.
clopidogrel and aspirin groups. Based on these find- In contrast to the consistent finding of a favorable
ings, clopidogrel has been approved for the reduc- benefit/risk profile of dual antiplatelet therapy in
tion of atherosclerotic events in patients with recent patients with acute coronary syndromes,253,256,257 the
stroke, recent MI, or established peripheral arterial same strategy was not proven successful when com-
disease. TTP can occur within the first 2 weeks after pared to clopidogrel alone in patients after a recent
the initiation of clopidogrel therapy.252 ischemic stroke or TIA,255 when compared to aspirin

alone in patients at high risk for atherothrombotic duces a functional thrombasthenic phenotype.267
events,258 or when compared to oral anticoagulation Approximately 40,000 antibody molecules bind to
in patients with atrial fibrillation.259 Although there the surface of platelets, but because they probably
might be mechanistic reasons underlying this appar- bind bivalently there are probably 80,000 GP-IIb/
ent heterogeneity in treatment effects, it is important IIIa receptors per platelet.260 Platelet aggregation is
to emphasize that the size of the additional benefit significantly inhibited at antibody doses that de-
associated with dual antiplatelet therapy vs aspirin crease the number of available receptors to ⬍ 50%
alone in patients with acute coronary syndromes is of normal. Platelet aggregation is nearly completely
only a fraction (about one third) of the benefit abolished at approximately 80% receptor blockade,
associated with aspirin vs no antiplatelet therapy. but the bleeding time is only mildly affected at this
Perhaps more importantly, both CURE253 and level of receptor blockade. It is only with ⬎ 90%
COMMIT257 investigators tested realistic hypotheses receptor blockade that the bleeding time becomes
of relative risk reduction (17% and 10%, respec- extremely prolonged.260 Because of concerns about
tively) and actually observed reductions (20% and immunogenicity of the original 7E3 antibody, a
9%, respectively) that were consistent with these mouse/human chimeric 7E3 Fab (abciximab) was
conservative expectations. In contrast, both the Man- created for clinical development.
agement of Atherothrombosis with Clopidogrel in Pharmacokinetic data on abciximab indicate that
High-Risk Patients (MATCH)255 and the Clopi- following IV bolus administration, free plasma con-
dogrel and High Atherothrombotic Risk and Isch- centrations decrease rapidly (initial half-life of about
emic Stabilization, Management, and Avoidance 30 min) as a result of rapid binding to platelet
(CHARISMA)258 investigators tested overoptimistic GP-IIb/IIIa receptors, with approximately 65% of
hypotheses of risk reduction and actually observed the injected antibody becoming attached to platelets
only a fraction (approximately one half to one third) in the circulation and spleen.268 After bolus injection
of the expected benefit. As expected, major bleeding of abciximab, a dose-dependent inhibition of ADP-
was increased by dual antiplatelet therapy in both induced platelet aggregation was recorded in pa-
MATCH255 and CHARISMA.258 tients judged to be at moderate to high risk of
percutaneous transluminal coronary angioplasty (PTCA)-
associated ischemic complications.268 A bolus dose of
5.0 Integrin ␣IIb␤3 (GP-IIb/IIIa) Receptor 0.25 mg/kg was found to result in blockade of ⬎ 80%
Antagonists of platelet receptors and reduce platelet aggregation
in response to 20 ␮mol/L ADP to ⬍ 20% of baseline.
Given the redundancy of discrete pathways lead-
A steep dose-response curve was apparent in this
ing to platelet aggregation, it is not surprising that
study.268 Peak effects on receptor blockade, platelet
the clinical efficacy of aspirin, ticlopidine, and clopi-
aggregation, and bleeding time were observed at the
dogrel is only partial. These drugs, while inhibiting
first sampling time of 2 h after bolus administration
TXA2- or ADP-mediated platelet aggregation, leave
of 0.25 mg/kg. Gradual recovery of platelet function
the activity of other platelet agonists, such as throm-
then occurred over time, with bleeding times return-
bin, largely unaffected. Following recognition that
ing to near normal values by 12 h.268 Platelet aggre-
the expression of functionally active integrin ␣IIb␤3
gation in response to 20 ␮mol/L ADP returns to
(GP-IIb/IIIa) on the platelet surface is the final
ⱖ 50% of baseline within 24 h in most patients, and
common pathway of platelet aggregation, regardless
within 48 h in nearly all patients. Small amounts of
of the initiating stimulus, this GP became the target
abciximab can be detected on circulating platelets as
of novel antiplatelet drugs.260 –263 The inhibitors of
late as 14 days after administration, presumably as a
GP-IIb/IIIa include monoclonal antibodies against
result of antibody redistribution from platelet to
the receptor, naturally occurring Arg-Gly-Asp se-
platelet.269
quence (RGD) containing peptides isolated from
The receptor blockade, inhibition of platelet
snake venoms, synthetic RGD- or Lys-Gly-Asp se-
aggregation, and prolongation of bleeding time
quence (KGD) containing peptides, and peptidomi-
produced by administering a 0.25 mg/kg bolus
metic and nonpeptide RGD mimetics that compete
dose of abciximab could be maintained for 12 h by
with fibrinogen, von Willebrand factor, and/or per-
administering a 10 ␮g/min infusion during that
haps other ligands for occupancy of the platelet
time period.268 This regimen was chosen for the
receptor.264 –266
phase III trial (Evaluation of 7E3 for the Prevention
of Ischemic Complications [EPIC])270 that demon-
5.1 Abciximab
strated the clinical efficacy of abciximab, added to
Blockade of GP-IIb/IIIa receptors by murine conventional antithrombotic therapy, in reducing the
monoclonal antibodies such as 7E3 essentially in- incidence of ischemic events in patients undergoing

PTCA (discussed in the “Antithrombotic Therapy for OR for major bleeding (1.18; p ⫽ 0.06) and essen-
Non-ST-Segment Elevation Acute Coronary Syn- tially no increase in the OR for minor bleeding (1.01;
dromes” chapter). Subsequently, the 10-␮g/min in- p ⫽ 0.94).284
fusion was modified to 0.125 ␮g/kg/min (to a maxi- Although the antiplatelet effect of abciximab in
mum of 10 ␮g/min) to adjust for differences in body preventing vascular occlusion by suppressing platelet
weight. aggregation is likely to be the major mechanism for
Reteplase and/or ticlopidine do not affect the its beneficial effects, it is quite possible that the
pharmacodynamics of abciximab.271 Pretreating potent inhibition of thrombus formation by this
platelets with tirofiban or eptifibatide does not alter antibody may result in decreased thrombin forma-
the subsequent binding of abciximab to platelets.272 tion.285 In fact, abciximab produced dose-dependent
It is unclear, however, whether abciximab can bind inhibition of tissue factor-induced thrombin genera-
simultaneously with either tirofiban or eptifibatide to tion, reaching a plateau of 45 to 50% inhibition at
a single GP-IIb/IIIa receptor, and indirect studies concentrations ⱖ 15 ␮g/mL.285 Whether the inhibi-
using monoclonal antibodies273 raise the possibility tion of thrombin generation by abciximab contrib-
that abciximab binding may decrease the binding of utes to its immediate antithrombotic effect remains
the other drugs.274 to be established. Abciximab is unique among the
Major bleeding was significantly increased in GP-IIb/IIIa antagonists in also blocking the ␣v␤3
abciximab-treated patients in EPIC.270 Subse- receptor at therapeutic doses260 and binding to an
quently, however, it was found that a reduction in activated form of the leukocyte ␣M␤2 receptor261,262;
the dosage of concomitant heparin and more rapid it is unclear whether any of the effects of abciximab
sheath removal could greatly reduce the bleeding are due to inhibition of these receptors.
complications attendant to abciximab administra-
tion.275 Besides hemorrhage, thrombocytopenia rep-
5.2 Tirofiban
resents an important side effect of abciximab treat-
ment. Approximately 1 to 2% of patients treated with Tirofiban (MK-383; Aggrastat) is a nonpeptide
abciximab have platelet counts ⬍ 50,000/␮L, of derivative of tyrosine that selectively inhibits the
which approximately 0.5 to 1% reflect very rapid GP-IIb/IIIa receptor, with minimal effects on the
(beginning within 2 h of administration) and severe ␣v␤3 vitronectin receptor.286,287 It inhibits platelet
(⬍ 20,000/␮L) decreases. The abciximab package aggregation of gel-filtered platelets induced by 10
insert specifies that a platelet count should be ob- ␮mol/L ADP with an inhibitory concentration (IC50)
tained 2 to 4 h after initiating therapy, thus permit- of 9 nM, but the IC50 for inhibition of human
ting the rapid identification of patients with throm- umbilical vein adhesion to vitronectin, which de-
bocytopenia developing. In almost all cases, the pends on ␣v␤3 vitronectin receptors, is 62 ␮mol/L.260
thrombocytopenia can be treated effectively by stop- Both renal and biliary excretion contribute to tirofi-
ping the drug and, if necessary, administering platelet ban clearance, with unchanged tirofiban found in
transfusions, with recovery occurring over several urine and feces.288
days.270,275,278 Binding of patient antibody to abcix- When administered to humans at 0.15 ␮g/kg/min
imab-treated platelets has been reported in patients for 4 h, tirofiban produced a 2.5-fold increase in
with abciximab-associated thrombocytopenia, but bleeding time and 97% inhibition of ADP-induced
the nature of the binding is unclear.279 Delayed platelet aggregation.289,290 The mean plasma clear-
thrombocytopenia has been ascribed to abciximab ance was 329 mL/min, and the half-life in plasma was
therapy, but its prevalence is unknown. In the EPIC 1.6 h. After stopping tirofiban therapy, bleeding
trial,270 approximately 6% of patients treated with times returned to normal within 4 h, and inhibition
abciximab developed antibodies to the variable re- of platelet aggregation declined to approximately
gion(s) of abciximab (human antichimeric antibody). 20%. When administered with aspirin, the bleeding
Few data are currently available to assess the poten- time increased 4.1 ⫾ 1.5-fold, even though tirofiban
tial risks of reinjecting abciximab,280 –282 which the- plasma levels were unaffected. The plasma concen-
oretically include anaphylaxis, neutralization of in- tration of tirofiban needed to inhibit platelet aggre-
jected abciximab, and thrombocytopenia. It appears, gation by 50% decreased, however, from approxi-
however, that the risk of thrombocytopenia is greater mately 12 ng/mL to approximately 9 ng/mL when
with abciximab readministration,282,283 especially if aspirin was coadministered. Peak plasma concentra-
the drug is readministered relatively soon after the tions were approximately 40 ng/mL, and the plasma
initial administration, and that the mechanism in- levels decreased to ⬍ 3 ng/mL within 6 h after
volves antibody binding to abciximab-coated plate- therapy was discontinued.
lets.279 In patients with chronic renal insufficiency, In a pilot study, 73 patients undergoing PTCA
abciximab produced only a modest increase in the were treated with aspirin, heparin, and bolus doses

of tirofiban of 5, 10, or 15 ␮g/kg followed by tirofiban in an overestimation of inhibition of platelet aggre-
infusions of 0.05, 0.10, and 0.15 ␮g/kg/min, respec- gation by eptifibatide298; thus, it is unclear whether
tively.291 The onset of platelet inhibition was rapid, there is a differential effect of eptifibatide on the
with platelet aggregation in response to 5 ␮mol/L bleeding time.
ADP inhibited by 93% and 96%, respectively, within A study299 of 14C-eptifibatide administered as a
5 min of administering the two higher-dose regi- single 135 ␮g/kg IV bolus revealed peak plasma
mens. Bleeding times at 2 h after starting the concentrations of 879 ⫾ 251 ng/mL (mean ⫾ SD) at
infusion were 19.5 min, ⬎ 30 min, and ⬎ 30 min, 5 min, a distribution half-life of 5 ⫾ 2.5 min, and a
respectively. At the end of the infusion (16 to 24 h), terminal elimination half-life of 1.1 ⫾ 0.17 h. Of the
platelet aggregation was inhibited by 57%, 87%, and approximately 73% of administered radioactivity re-
95%, respectively, in response to the escalating covered in 72 h, renal clearance accounted for 98%
tirofiban regimens. Platelet aggregation began to of total recovered radioactivity and approximately
return toward normal within 1.5 h after discontinu- 40% of total body clearance. Unmodified eptifi-
ing the infusion in all groups; 4 h after discontinuing batide, deamidated eptifibatide, and more polar
therapy, platelet aggregation inhibition decreased to metabolites were all found in the urine, but only
⬍ 50%, even in the group receiving the highest dose. trace amounts of radioactivity were found in the
In patients with renal insufficiency (creatinine breath and feces.
clearance ⬍ 30 mL/min), plasma clearance of tirofi- Because renal clearance is an important compo-
ban was reduced, and plasma half-life increased by nent of eptifibatide catabolism, patients with renal
more than threefold.292 The manufacturer recom- impairment can have prolonged inhibition of platelet
mends reducing both the bolus and infusion doses by function after receiving eptifibatide. This is of par-
50%, but the pharmacokinetic basis for this recom- ticular theoretical concern because patients with
mendation has been challenged.292 In the Platelet end-stage renal failure have platelet dysfunction.
Receptor Inhibition in Ischemic Syndrome Manage- The proper dose of eptifibatide in patients with
ment in Patients Limited by Unstable Signs and modest-to-moderate renal insufficiency (creatinine,
Symptoms trial,293 the 40 patients with creatinine 2 to 4 mg/dL) is uncertain.292 In the ESPRIT,300
clearance ⬍ 30 mL/min were found to have in- patients with creatinine clearances ⱕ 60 mL/min
creased risk of bleeding, and tirofiban treatment had increased major and minor bleeding rates com-
further increased the risk. pared to those with creatinine clearances ⬎ 60 mL/
Severe, but reversible thrombocytopenia has been min, and eptifibatide treatment increased both major
reported278,294 in a small percentage of patients and minor bleeding in both groups of patients.
treated with tirofiban; an immunologic mechanism Because the steady-state level of eptifibatide is
has been proposed, mediated by preformed antibod- approximately 1,900 ng/mL when using an infusion
ies to a conformation of the GP-IIb/IIIa receptor rate of 2 ␮g/kg/min,301 the ratio of eptifibatide
induced by the binding of tirofiban to the receptor. molecules to GP-IIb/IIIa molecules is ⬎ 50:1302;
To our knowledge, no data are available on the safety thus, platelet transfusions may not be able to reverse
of reinfusing tirofiban, but high-titer antibodies have the effects of the drug, although in vitro data raise
been identified in patients with thrombocytopenia some hope in this regard.303 Treatment with eptifi-
after repeat administration.295 batide prolongs the activated clotting time of hepa-
rinized patients, suggesting an inhibitory effect on
thrombin generation.304,305
5.3 Eptifibatide
In 21 patients undergoing elective PTCA or direc-
Eptifibatide (Integrilin; Millenium Pharmaceuti- tional coronary atherectomy who were treated with
cals, Schering Corporation; Kenilworth, NJ) is a aspirin, heparin (10,000 U bolus ⫹ additional doses
synthetic disulfide-linked cyclic heptapeptide. It is to maintain an activated clotting time at 300 to
patterned after the KGD sequence found in the 350 s), and a bolus dose of 90 ␮g/kg of eptifibatide
snake venom disintegrin obtained from Sistrurus followed by a 1-␮g/kg/min infusion for 4 or 12 h,
miliarius barbouri (barbourin) and has high, but not platelet aggregation was measured before, 1 h after
absolute, specificity for inhibition of GP-IIb/IIIa the bolus, at the end of the infusion, and 4 h after the
compared with inhibition of the ␣v␤3 vitronectin end of the infusion.304 The extent of platelet aggre-
receptor.296,297 Preliminary reports suggested that gation in response to 20 ␮mol/L ADP decreased
eptifibatide produced less prolongation of the bleed- from approximately 80% before eptifibatide to ap-
ing time than other GP-IIb/IIIa inhibitors at doses proximately 15% both at 1 h after the bolus dose and
producing comparable inhibition of platelet aggrega- at the end of the infusion. There was, however,
tion. Later studies found that the citrate anticoagu- significant interindividual variation in the inhibitory
lation used for platelet aggregation studies resulted responses (95% CI, 0% to approximately 30%, and

0% to approximately 40%, respectively) at the two tions of different single bolus doses (␮g/kg) followed
time points tested. Four hours after stopping the by different infusion doses (␮g/kg/min) [135/0.75,
infusion, the average aggregation response was ap- 180/2.0, 250/3.0] were evaluated in patients with
proximately 55%, but there was marked individual acute coronary syndromes307 and during PCI.301,304,306
variation (95% CI, approximately 10% to 90%). High-level inhibition of ADP-induced aggregation
Median bleeding times were prolonged with eptifi- could be achieved soon after the bolus dose, but
batide therapy, going from approximately 6 min there was an early loss of inhibition of platelet
before treatment to approximately 26 min both at 1 h aggregation before steady-state was achieved.
after beginning the infusion and at the end of the Regimens using a second bolus dose 30 min after
infusion. The bleeding times returned to normal the first were then studied (180/2.0 plus second
(median, 15 min) within 15 min after stopping bolus of 90; 250/2.0 plus second bolus of 125)301
eptifibatide therapy and declined to approximately From these studies, the dose used in the ES-
12 min after stopping the drug therapy for 1 h. At PRIT300 (bolus dose of 180 ␮g/kg/min, second
each time point, however, there were considerable bolus dose of 180 ␮g/kg at 10 min followed by 2
interindividual differences.304 ␮g/kg/min infusion) was selected.301
In a later study,306 four eptifibatide regimens were A modest increase in hemorrhagic complications
tested in 54 patients undergoing coronary interven- has been reported in patients treated with eptifi-
tions who also were treated with aspirin and heparin: batide in the PURSUIT trial308 and the ES-
(1) 180 ␮g/kg bolus ⫹ 1 ␮g/kg/min infusion for 18 to PRIT.300,309,310 Eptifibatide treatment has been
24 h (n ⫽ 4); (2) 135 ␮g/kg bolus ⫹ 0.5 ␮g/kg/min associated with a small increase in profound
infusion for 18 to 24 h (n ⫽ 16); (3) 90 ␮g/kg thrombocytopenia.308,311 An immunologic mecha-
bolus ⫹ 0.75 ␮g/kg/min infusion for 18 to 24 h nism has been identified in some patients.295 Thus,
(n ⫽ 6); and (4) 135 ␮g/kg bolus ⫹ 0.75 ␮g/kg/min patients receiving eptifibatide should be moni-
for 18 to 24 h (n ⫽ 28). Fifteen minutes after the 180 tored soon after initiation of therapy for develop-
␮g/kg bolus dose, platelet aggregation was inhibited ment of thrombocytopenia. An algorithm for the
by ⬎ 95% in response to 20 ␮mol/L ADP, with detection and management of thrombocytopenia
virtually no interindividual variation, whereas the after GP-IIb/IIIa blockade has been proposed.278 To
135-␮g/kg bolus dose resulted in 80 to 90% inhibi- our knowledge, no data are available about the safety
tion in 75% of the patients, and the 90 ␮g/kg bolus of reinfusing eptifibatide, but high levels of antibody
produced only slightly less inhibition than the 135 that bind to platelets in the presence of eptifibatide
␮g/kg dose. The inhibition of platelet aggregation have been found in patients who have thrombocyto-
achieved with the 180 ␮g/kg bolus dose was sus- penia after reexposure to eptifibatide.295
tained throughout the infusion by the 1 ␮g/kg/min
dose, but there was a tendency for the platelet
5.4 Efficacy and Safety of IV GP-IIb/IIIa
aggregation response to return toward normal during
Antagonists
the infusion in some patients given the 0.75 ␮g/kg/
min dose, and the return of the platelet aggregation The efficacy and safety of GP-IIb/IIIa antagonists
response toward normal was more marked in those have been evaluated initially in patients undergoing
given the 0.5 ␮g/kg/min infusion dose. Two hours PCI. More than 20,000 patients have been enrolled
after discontinuing the eptifibatide infusion, there in nine studies of abciximab, eptifibatide, and tirofi-
was substantial return of platelet function in all ban. The first of these phase III trials, the EPIC
groups and return of more than half of the baseline trial,270 resulted in approval in many countries of
aggregation response in all groups after 4 h. Median abciximab (ReoPro; Eli Lilly Company and Cento-
bleeding times were prolonged in all groups at the cor; Indianapolis, IN) in 1994 for PCI patients at
time the infusion was terminated (22, 12, 12, and 17 high risk for ischemic complication. Eptifibatide has
min, respectively, compared with control values of 7 been studied in the IMPACT-II305 and ESPRIT
to 8 min), and they returned toward normal after 1 h trials,309,310 and tirofiban has been studied in the
(9, 10, 9, and 11 min, respectively). As in the RESTORE trial.312 Although neither the IMPACT-II nor
previous study, activated clotting times were longer the RESTORE trials achieved their predefined effi-
in patients treated with eptifibatide plus heparin cacy end points, there was a positive trend in each
than in those treated with placebo plus heparin. case. Eptifibatide received approval from the FDA for
After the effect of citrate was discovered, the dose of PCI in 1998 based on data from the IMPACT-II and
eptifibatide was increased, and platelet studies were PURSUIT trials, and the dosing was modified based on
conducted on blood anticoagulated with the direct the efficacy demonstrated in the ESPRIT.309,310 The
thrombin inhibitor D-Phe-Pro-Arg chloromethyl ke- CAPTURE trial313 demonstrated the efficacy of an
tone, which does not chelate calcium.301 Combina- 18- to 24-h abciximab treatment before PCI in

patients with unstable angina refractory to conven- efficacy and safety of tirofiban, lamifiban (a nonpep-
tional antithrombotic and antianginal therapy. The tide GP-IIb/IIIa blocker whose development has
EPILOG trial275 demonstrated the efficacy of abcix- been discontinued), eptifibatide, and abciximab in
imab in a broad patient population undergoing PCI, approximately 25,000 patients with acute coronary
not just high-risk patients as enrolled in the EPIC syndromes without persistent ST-segment elevation
and CAPTURE trials. The EPISTENT trial demon- randomized to receive a GP-IIb/IIIa antagonist or
strated that abciximab decreases the frequency of placebo in addition to conventional antithrombotic
ischemic complications of PCI associated with stent therapy. These studies demonstrated a 0 to 27%
insertion during the first 30 days, and that there are relative risk reduction in MI or death at 30 days.
fewer ischemic complications during this period in Both eptifibatide and tirofiban have received ap-
patients treated with PCI and abciximab alone with- proval from the FDA for the treatment of acute
out stent insertion vs those treated with stent coronary syndromes, including patients who are to
alone.314 Furthermore, the 1-year mortality differ- be managed medically and those undergoing PCI.
ence was statistically significant between stent alone However, in the GUSTO IV-ACS trial,326 abciximab
(2.4%) and stent plus abciximab (1%), and this for 24 h (0.25 mg/kg bolus followed by a 0.125
mortality difference was sustained for longer peri- ␮g/kg/min infusion) or 48 h was not beneficial as
ods.315 Both abciximab and stenting were studied in first-line medical treatment in patients with acute
the CADILLAC trial316 of patients with MI. In this coronary syndromes. A metaanalysis of all major
group of patients, who appeared to be at relative low randomized clinical trials of GP-IIb/IIIa antagonists
risk, abciximab had a beneficial effect in the PTCA in patients with acute coronary syndromes who were
group but did not affect death or reinfarction in the not routinely scheduled to undergo early coronary
stent group.316 revascularization suggests a 9% reduction in the odds
During the past 4 years, a series of randomized of death or MI at 30 days.327 However, the true size
clinical trials conducted by the Intracoronary Stent- of the additional benefit resulting from short-term,
ing and Antithrombotic Regimen Group317–320 have high-grade blockade of GP-IIb/IIIa combined with
reexamined the efficacy and safety of GP-IIb/IIIa standard antithrombotic therapy is somewhat uncer-
blockade in a broad range of patients undergoing tain because the 95% CI ranged from 2 to 16%.
PCI on dual oral antiplatelet therapy. Using a Moreover, the 1% absolute difference in death or MI
600-mg clopidogrel loading dose given at least 2 h was balanced by an absolute excess of 1% in major
before PCI in all patients, the investigators evalu- bleeding complications associated with GP-IIb/IIIa
ated the effects of adjunctive abciximab in low- to antagonists vs control.327
intermediate-risk patients,317 in patients undergo- Thus, the benefit/risk profile of currently available
ing revascularization of small-diameter vessels,318 in GP-IIb/IIIa antagonists is substantially uncertain for
patients with diabetes mellitus,319 and in patients patients with acute coronary syndromes who are not
with non–ST-segment elevation acute coronary syn- routinely scheduled for early revascularization. In
dromes.320 In stable patients undergoing elective contrast, for high-risk patients undergoing PCI, in-
PCI, pretreatment with 600 mg of clopidogrel pro- tensification of antiplatelet therapy by adding an IV
vides platelet inhibition sufficient to enable a safe GP-IIb/IIIa blocker is an appropriate strategy to
procedure without the need of GP-IIb/IIIa block- reduce the risk of procedure-related thrombotic
ade.317–319 However, the same abciximab regimen complications.
was associated with a statistically significant 25% Phase II trials in acute MI with abciximab, eptifi-
relative risk reduction in the 30-day combined end batide, and lamifiban have suggested potential ben-
point of death, MI, or urgent target vessel revascu- efits of GP-IIb/IIIa blockade as an adjunct to throm-
larization in patients with acute coronary syn- bolysis. The Thrombolysis in Myocardial Infarction
dromes.320 Although the additional benefit of GP- 14A trial328 demonstrated that combining abciximab
IIb/IIIa blockade appeared to be confined to with aspirin and reduced-dose tissue plasminogen
patients with an elevated troponin level (⬎ 0.03 activator in the treatment of acute MI resulted in
␮g/L), the p value for the interaction was not improved TIMI 3 flow rates at 60 min and 90 min
statistically significant. after starting therapy compared to the best full-dose
Abciximab was compared to tirofiban as treatment tissue plasminogen activator regimen. The bleeding
for PCI in the TARGET study.321,322 Abciximab risk was not substantially increased.328 The GUSTO
treatment was found to be associated with a statisti- V trial329 compared the efficacy and safety of half-
cally significant lower rate of ischemic complications dose reteplase and full-dose abciximab vs standard-
after 30 days; at 6 months, the differences were less dose reteplase in 16,588 patients in the first 6 h of
apparent. evolving ST-segment elevation MI. The primary end
Five completed trials308,323–326 have examined the point of 30-day mortality was similar in the two

treatment groups (5.6% vs 5.9%). Combination ther- 10 Burch J, Stanford P, Majerus P. Inhibition of platelet
apy led to a consistent reduction in secondary com- prostaglandin synthetase by oral aspirin. J Clin Invest 1979;
61:314 –319
plications of MI, including reinfarction, which was
11 Majerus P. Arachidonate metabolism in vascular disorders.
partly counterbalanced by increased extracranial J Clin Invest 1983; 72:1521–1525
bleeding.329 There was no mortality benefit of com- 12 Smith W, Garavito R, DeWitt D. Prostaglandin endoperox-
bined therapy after 1 year, and thus there appears to ide H synthases (cyclooxygenases)-1 and -2. J Biol Chem
be little or no net benefit in combined therapy.330 1996; 271:33157–33160
Despite reassuring data from a phase II trial of 13 Loll P, Picot D, Garavito R. The structural basis of aspirin
abciximab in patients with acute ischemic stroke,331 a activity inferred from the crystal structure of inactivated
prostaglandin H2 synthase. Nat Struct Biol 1995; 2:637– 643
phase III trial has been stopped because of safety
14 Bala M, Chin C, Logan A, et al. Acetylation of prostaglandin
concerns. H2 synthases by aspirin is inhibited by redox cycling of the
peroxidase. Biochem Pharmacol 2008; 75:1472–1481
15 Clarke R, Mayo G, Price P, et al. Suppression of thrombox-
Conlict of Interest Disclosures ane A2 but not systemic prostacyclin by controlled-release
aspirin. N Engl J Med 1991; 325:1137–1141
Dr. Patrono discloses that he has received grant monies from 16 McAdam B, Catella-Lawson F, Mardini I, et al. Systemic
the University of Rome ‘La Sapienza,’ Catholic University School biosynthesis of prostacyclin by cyclooxygenase (COX)-2: the
of Medicine, the Italian Ministry of Research and University, the human pharmacology of a selective inhibitor of COX-2. Proc
European Commission, Bayer Italy, and MSD Italy. Dr. Patrono Natl Acad Sci U S A 1999; 96:272–277
has received consultant fees from Servier and NicOx. He has 17 Topper J, Cai J, Falb D, et al. Identification of vascular
served on the speakers bureau of AstraZeneca, Bayer AG, Eli endothelial genes differentially responsive to fluid mechan-
Lilly, Sanofi-Aventis, and Schering-Plough. ical stimuli: cyclooxygenase-2, manganese superoxide dis-
Dr. Hirsh discloses that he has received partial support for mutase, and endothelial cell nitric oxide synthase are selec-
writing two books, one on fondaparinux and one on low-molec- tively up-regulated by steady laminar shear stress. Proc Natl
ular-weight heparin. Acad Sci U S A 1996; 93:10417–10422
Dr. Roth reveals no real or potential conflicts of interest or 18 FitzGerald G, Oates J, Hawiger J, et al. Endogenous
commitment. biosynthesis of prostacyclin and thromboxane and platelet
Dr. Baigent discloses that he received grant monies from the function during chronic administration of aspirin in man.
British Heart Foundation for A Study of Cardiovascular Events in J Clin Invest 1983; 71:678 – 688
Diabetes (or ASCEND; for which he was a coinvestigator) and 19 Murata T, Ushikubi F, Matsuoka T, et al. Altered pain
served as an unpaid member of the AstraZeneca board for study perception and inflammatory response in mice lacking pros-
AZD6140. tacyclin receptor. Nature 1997; 388:678 – 682
20 Kearney P, Baigent C, Godwin J, et al. Do selective
cyclo-oxygenase-2 inhibitors and traditional non-steroidal
anti-inflammatory drugs increase the risk of atherothrombo-
References sis? Meta-analysis of randomized trials. BMJ 2006; 332:
1 Davi G, Patrono C. Platelet activation and atherothrombosis. 1302–1308
N Engl J Med 2007; 357:2482–2494 21 Grosser T, Fries S, FitzGerald G. Biological basis for the
2 Patrono C, Coller B, FitzGerald G, et al. Platelet-active cardiovascular consequences of COX-2 inhibition: therapeu-
drugs: the relationships among dose, effectiveness, and side tic challenges and opportunities. J Clin Invest 2006; 116:
effects. Chest 2004; 126(suppl):234S–264S 4 –15
3 Kaushansky K. Regulation of megakaryopoiesis. In: Loscalzo 22 Pedersen A, FitzGerald G. Dose-related kinetics of aspirin:
J, Schaefer AL, eds. Thrombosis and hemorrhage. Balti- presystemic acetylation of platelet cyclo-oxygenase. N Engl
more, MD: Williams & Wilkins, 1998 J Med 1984; 311:1206 –1211
4 Rocca B, Secchiero P, Ciabattoni G, et al. Cyclooxygenase-2 23 Cox D, Maree A, Dooley M, et al. Effect of enteric coating
expression is induced during human megakaryopoiesis and on antiplatelet activity of low-dose aspirin in healthy volun-
characterizes newly formed platelets. Proc Natl Acad Sci teers. Stroke 2006; 37:2153–2158
U S A 2002; 99:7634 –7639 24 McAdam B, Keimowitz R, Maher M, et al. Transdermal
5 Lindemann S, Tolley N, Dixon D, et al. Activated platelets modification of platelet function: an aspirin patch system
mediate inflammatory signaling by regulated interleukin 1␤ results in marked suppression of platelet cyclooxygenase.
synthesis. J Cell Biol 2001; 154:485– 490 J Pharmacol Exp Ther 1996; 277:559 –564
6 Patrono C, Garcı́a Rodrı́guez L, Landolfi R, et al. Low-dose 25 Patrono C, Ciabattoni G, Patrignani P, et al. Clinical phar-
aspirin for the prevention of atherothrombosis. N Engl macology of platelet cyclooxygenase inhibition. Circulation
J Med 2005; 353:2373–2383 1985; 72:1177–1184
7 Antithrombotic Trialists Collaboration. Collaborative meta- 26 Demers L, Budin R, Shaikh B. The effects of aspirin on
analysis of randomized trials of antiplatelet therapy for megakaryocyte prostaglandin production. Proc Soc Exp Biol
prevention of death, myocardial infarction, and stroke in Med 1980; 163:24 –29
high-risk patients. BMJ 2002; 324:71– 86 27 Burch J, Baenziger N, Stanford N, et al. Sensitivity of fatty
8 Roth G, Majerus P. The mechanism of the effect of aspirin acid cyclo-oxygenase from human aorta to acetylation by
on human platelets: I. Acetylation of a particulate fraction aspirin. Proc Natl Acad Sci U S A 1978; 75:5181–5184
protein. J Clin Invest 1975; 56:624 – 632 28 Burch J, Majerus P. The role of prostaglandins in platelet
9 Roth G, Stanford N, Majerus P. Acetylation of prostaglandin function. Semin Hematol 1979; 16:196 –207
synthase by aspirin. Proc Natl Acad Sci U S A 1975; 29 Cerskus A, Ali M, Davies B, et al. Possible significance of
72:3073–3077 small numbers of functional platelets in a population of

aspirin-treated platelets in vitro and in vivo. Thromb Res effects on platelet aggregation and thromboxane formation.
1980; 18:389 –397 Lancet 1984; 1:1261–1264
30 O’Brien J. Effects of salicylates on human platelets. Lancet 48 Goldman S, Copeland J, Moritz T, et al. Improvement in
1968; 1:779 –783 early saphenous vein graft patency after coronary artery
31 Reilly I, FitzGerald G. Inhibition of thromboxane formation bypass surgery with antiplatelet therapy: results of a Veter-
in vivo and ex vivo: implications for therapy with platelet ans Administration cooperative study. Circulation 1988;
inhibitory drugs. Blood 1987; 69:180 –186 77:1324 –1332
32 RISC Group. Risk of myocardial infarction and death during 49 Brown B, Cukingnan R, DeRouen T, et al. Improved graft
treatment with low dose aspirin and intravenous heparin in patency in patients treated with platelet-inhibiting therapy
men with unstable coronary artery disease. Lancet 1990; after coronary bypass surgery. Circulation 1985; 72:138 –146
336:827– 830 50 Turpie A, Gent G, Laupacis A, et al. A comparison of aspirin
33 Juul-Moller S, Edvardsson N, Jahnmatz B, et al. Double- with placebo in patients treated with warfarin after heart
valve replacement. N Engl J Med 1993; 329:524 –529
blind trial of aspirin in primary prevention of myocardial
51 Altman R, Boullon F, Rouvier J, et al. Aspirin and prophy-
infarction in patients with stable chronic angina pectoris.
laxis of thromboembolic complications in patients with
Lancet 1992; 340:1421–1425
substitute heart valves. J Thorac Cardiovasc Surg 1976;
34 SALT Collaborative Group. Swedish Aspirin Low-Dose 72:127–129
Trial (SALT) of 75 mg aspirin as secondary prophylaxis after 52 Chesebro J, Fuster V, Elveback L, et al. Trial of combined
cerebrovascular ischemic events. Lancet 1991; 338:1345– warfarin plus dipyridamole or aspirin therapy in prosthetic
1491 heart valve replacement: danger of aspirin compared with
35 Lindblad B, Persson N, Takolander R, et al. Does low-dose dipyridamole. Am J Cardiol 1983; 51:1537–1541
acetylsalicylic acid prevent stroke after carotid surgery? A 53 Dale J, Myhre E, Storstein O, et al. Prevention of arterial
double-blind, placebo-controlled randomized trial. Stroke thromboembolism with acetylsalicylic acid: a controlled
1993; 24:1125–1128 clinical study in patients with aortic ball valves. Am Heart J
36 Diener H, Cunha L, Forbes C, et al. European Stroke 1977; 94:101–111
Prevention Study: II. Dipyridamole and acetylsalicylic acid 54 Harter H, Burch J, Majerus P, et al. Prevention of throm-
in the secondary prevention of stroke. J Neurol Sci 1996; bosis in patients on hemodialysis by low-dose aspirin. N Engl
143:1–13 J Med 1979; 301:577–579
37 Landolfi R, Marchioli R, Kutti J, et al. Efficacy and safety of 55 ISIS-2 Collaborative Group. Randomized trial of intrave-
low-dose aspirin in polycythemia vera. N Engl J Med 2004; nous streptokinase, oral aspirin, both or neither among
350:114 –124 17,187 cases of suspected acute myocardial infarction:
38 McKenna R, Galante J, Bachmann F, et al. Prevention of ISIS-2. Lancet 1988; 2:349 –360
venous thrombo-embolism after total knee replacement by 56 Bousser M, Eschwege E, Haugenau M, et al. ‘AICLA’
high-dose aspirin or intermittent calf and thigh compression. controlled trial of aspirin and dipyridamole in the secondary
BMJ 1980; 280:514 –517 prevention of athero-thrombotic cerebral ischemia. Stroke
39 Harris W, Athanasoulis C, Waltman A, et al. High and 1983; 14:5–14
low-dose aspirin prophylaxis against venous thromboembolic 57 Barnett H, Gent M, Sackett D, et al. A randomized trial of
disease in total hip replacement. J Bone Joint Surg Am 1982; aspirin and sulfinpyrazone in threatened stroke: the Cana-
64:63– 66 dian Cooperative Study Group. N Engl J Med 1978; 299:
40 Harris W, Athanasoulis C, Waltman A, et al. Prophylaxis of 53–59
deep vein thrombosis after total hip replacement: dextran 58 Fields W, Lemak N, Frankowski R, et al. Controlled trial of
and external compression compared with 1.2 or 0.3 gram of aspirin in cerebral ischemia. Stroke 1977; 8:301–314
aspirin daily. J Bone Joint Surg Am 1985; 67:57– 62 59 International Stroke Trial Collaborative Group. The Inter-
41 Farrel B, Godwin J, Richards S, et al. The United Kingdom national Stroke Trial (IST): a randomized trial of aspirin,
transient ischemic attack (UK-TIA) aspirin trial: final results. subcutaneous heparin, both, or neither among 19,435 pa-
J Neurol Neurosurg Psychiatry 1991; 54:1044 –1054 tients with acute ischemic stroke. Lancet 1997; 349:1569 –
42 Dutch TIA Trial Study Group. A comparison of two doses of 1581
aspirin (30 mg vs 283 mg a day) in patients after a transient 60 CAST (Chinese Acute Stroke Trial) Collaborative Group.
ischemic attack or minor ischemic stroke. N Engl J Med CAST: randomized placebo-controlled trial of early aspirin
1991; 325:1261–1266 use in 20,000 patients with acute ischemic stroke. Lancet
43 Taylor D, Barnett H, Haynes R, et al. Low-dose and 1997; 349:1641–1649
high-dose acetylsalicylic acid for patients undergoing carotid 61 Tartaglia A, Goldberg J, Berk P, et al. Adverse effects of anti
endarterectomy: a randomized controlled trial; ASA and aggregating platelet therapy in the treatment of polycythe-
Carotid Endarterectomy (ACE) Trial Collaborators. ASA mia vera. Semin Hematol 1986; 23:172–176
and Carotid Endarterectomy (ACE) Trial Collaborators. 62 Patrignani P, Filabozzi P, Patrono C. Selective cumulative
Lancet 1999; 353:2179 –2184 inhibition of platelet thromboxane production by low-dose
44 Lewis H, Davis J, Archibald D, et al. Protective effects of aspirin in healthy subjects. J Clin Invest 1982; 69:1366 –1372
aspirin against acute myocardial infarction and death in men 63 Peters R, Mehta S, Fox K, et al. Effects of aspirin dose when
with unstable angina: results of a Veterans Administration used alone or in combination with clopidogrel in patients
cooperative study. N Engl J Med 1983; 309:396 – 403 with acute coronary syndromes: observations from the Clo-
45 Theroux P, Ouimet H, McCans J, et al. Aspirin, heparin, or pidogrel in Unstable angina to prevent Recurrent Events
both to treat acute unstable angina. N Engl J Med 1988; (CURE) study. Circulation 2003; 108:1682–1687
319:1105–1111 64 Hanson S, Harker L, Bjomssorl T. Effect of platelet-modifying
46 Cairns J, Gent M, Singer J, et al. Aspirin, sulfinpyrazone, or drugs on arterial thromboembolism in baboons: aspirin poten-
both in unstable angina. N Engl J Med 1985; 313:1369 –1375 tiates the antithrombotic actions of dipyridamole and sulfin-
47 Lorenz R, Weber M, Kotzur J, et al. Improved aortocoro- pyrazone by mechanism(s) independent of platelet cyclooxy-
nary bypass patency by low-dose aspirin (100 mg daily): genase inhibition. J Clin Invest 1985; 75:1591–1599

65 Harker L, Fuster V. Pharmacology of platelet inhibitors. aspirin is less than optimal in patients with vascular disease
J Am Coll Cardiol 1986; 8:21B–32B due to prothrombotic effects of erythrocytes on platelet
66 Buchanan M, Rischke J, Hirsh J. Aspirin inhibits platelet reactivity. Circulation 1998; 97:350 –355
function independent of the acetylation of cyclo-oxygenase. 87 Podhaisky H, Abate A, Polte T, et al. Aspirin protects
Thromb Res 1982; 25:363–373 endothelial cells from oxidative stress: possible synergism
67 Gaspari F, Viganò G, Grisio S, et al. Aspirin prolongs with vitamin E. FEBS Lett 1997; 417:349 –351
bleeding time in uremia by a mechanism distinct from 88 Husain S, Andrews N, Mulcahy D, et al. Aspirin improves
platelet cyclo-oxygenase inhibition. J Clin Invest 1987; 79: endothelial dysfunction in atherosclerosis. Circulation 1998;
1788 –1797 97:716 –720
68 Ratnatunga C, Edmondson S, Rees G, et al. High-dose 89 Roth G, Calverley D. Aspirin, platelets and thrombosis:
aspirin inhibits shear-induced platelet reaction involving theory and practice. Blood 1994; 83:885– 898
thrombin generation. Circulation 1992; 85:1077–1082 90 Patrono C. Aspirin resistance: definition, mechanisms and
69 Bjornsson T, Schneider D, Berger H. Aspirin acetylates clinical read-outs. J Thromb Haemost 2003; 1:1710 –1713
fibrinogen and enhances fibrinolysis: fibrinolytic effect is 91 Helgason C, Hoff J, Kondos G, et al. Platelet aggregation in
independent of changes in plasminogen activator levels. patients with atrial fibrillation taking aspirin or warfarin.
J Pharmacol Exp Ther 1988; 250:154 –160 Stroke 1993; 24:1458 –1461
70 Moroz L. Increased blood fibrinolytic activity after aspirin 92 Helgason C, Tortorice K, Winkler S, et al. Aspirin response
ingestion. N Engl J Med 1977; 296:525–529 and failure in cerebral infarction. Stroke 1993; 24:345–350
71 Green D, Davies R, Holmes G, et al. Fibrinolytic activity 93 Helgason C, Bolin K, Hoff J, et al. Development of aspirin
after administration of diflunisal and aspirin. Haemostasis resistance in persons with previous ischemic stroke. Stroke
1983; 13:394 –398 1994; 25:2331–2336
72 Quick A, Cleasceri L. Influence of acetylsalicylic acid and 94 Weksler B, Kent J, Rudolph D, et al. Effects of low dose
salicylamide on the coagulation of blood. J Pharmacol Exp aspirin on platelet function in patients with recent cerebral
Ther 1960; 128:95–99 ischemia. Stroke 1985; 16:5–9
73 Loew D, Vinazzer H. Dose-dependent influence of acetyl- 95 Mueller M, Salat A, Stangl P, et al. Variable platelet
salicylic acid on platelet functions and plasmatic coagulation response to low-dose ASA and the risk of limb deterioration
factors. Haemostasis 1976; 5:239 –249 in patients submitted to peripheral arterial angioplasty.
74 Kessels H, Beguin S, Andreee H, et al. Measurement of Thromb Haemost 1997; 78:1003–1007
thrombin generation in whole blood: the effect of heparin 96 Buchanan M, Brister S. Individual variation in the effects of
and aspirin. Thromb Haemost 1994; 72:78 – 83 ASA on platelet function: implications for the use of ASA
75 Szczeklik A, Krzanowski M, Gora P, et al. Antiplatelet drugs clinically. Can J Cardiol 1995; 11:221–227
and generation of thrombin in clotting blood. Blood 1992; 97 Gum P, Kottke-Marchant K, Poggio E, et al. Profile and
80:2006 –2011 prevalence of aspirin resistance in patients with cardiovas-
76 Roth G, Machuga E, Ozols J. Isolation and covalent struc- cular disease. Am J Cardiol 2001; 88:230 –235
ture of the aspirin-modified, active-site region of prostaglan- 98 Sane D, McKee S, Malinin A, et al. Frequency of aspirin
din synthetase. Biochemistry 1983; 22:4672– 4675 resistance in patients with congestive heart failure treated
77 Pinckard R, Hawkins D, Farr R. In vitro acetylation of with antecedent aspirin. Am J Cardiol 2002; 90:893– 895
plasma proteins, enzymes and DNA by aspirin. Nature 1969; 99 Berglund U, Wallentin L. Persistent inhibition of platelet
219:68 – 69 function during long-term treatment with 75 mg acetylsali-
78 Szczeklik A, Musial J, Undas A, et al. Inhibition of thrombin cylic acid daily in men with unstable coronary artery disease.
generation by aspirin is blunted in hypercholesterolemia. Eur Heart J 1991; 12:428 – 433
Arterioscler Thromb Vasc Biol 1996; 16:948 –954 100 Schwartz K, Schwartz D, Ghosheh K, et al. Compliance as a
79 North American Symptomatic Carotid Endarterectomy critical consideration in patients who appear to be resistant
Trial Collaborators. Beneficial effect of carotid endarterec- to aspirin after healing of myocardial infarction. Am J
tomy in symptomatic patients with high-grade carotid ste- Cardiol 2005; 95:973–975
nosis. N Engl J Med 1991; 325:445– 453 101 Lev E, Patel R, Maresh K, et al. Aspirin and clopidogrel
80 Barnett H, Kaste M, Meldrum H, et al. Aspirin dose in drug response in patients undergoing percutaneous coro-
stroke prevention: beautiful hypotheses slain by ugly facts. nary intervention: the role of dual drug resistance. J Am Coll
Stroke 1996; 27:588 –592 Cardiol 2006; 47:27–33
81 Steering Committee of the Physicians’ Health Study Re- 102 Carney R, Freedland K, Eisen S, et al. Adherence to a
search Group. Final report on the aspirin component of the prophylactic medication regimen with symptomatic versus
ongoing Physicians’ Health Study. N Engl J Med 1989; asymptomatic ischemic heart disease. Behav Med 1998;
321:129 –135 24:35–39
82 Ridker P, Cushman M, Stampfer M, et al. Inflammation, 103 Grotemeyer K, Scharafinski H, Husstedt I. Two-year
aspirin, and the risk of cardiovascular disease in apparently follow-up of aspirin responder and aspirin non-responder: a
healthy men. N Engl J Med 1997; 336:973–979 pilot-study including 180 post-stroke patients. Thromb Res
83 FitzGerald G, Patrono C. The coxibs, selective inhibitors of 1993; 71:397– 403
cyclooxygenase-2. N Engl J Med 2001; 345:433– 442 104 Bornstein N, Karepov V, Aronovich B, et al. Failure of
84 López-Farré A, Caramelo C, Esteban A, et al. Effects of aspirin treatment after stroke. Stroke 1994; 25:275–277
aspirin on platelet-neutrophil interactions: role of nitric 105 Chamorro A, Escolar G, Revilla M, et al. Ex vivo response to
oxide and endothelin-1. Circulation 1995; 91:2080 –2088 aspirin differs in stroke patients with single or recurrent
85 Santos M, Valles J, Aznar J, et al. Prothrombotic effects of events: a pilot study. J Neurol Sci 1999; 171:110 –114
erythrocytes on platelet reactivity: reduction by aspirin. 106 Gum P, Kottke-Marchant K, Welsh P, et al. A prospective,
Circulation 1997; 95:63– 68 blinded determination of the natural history of aspirin
86 Valles J, Santos M, Aznar J, et al. Erythrocyte promotion of resistance among stable patients with cardiovascular disease.
platelet reactivity decreases the effectiveness of aspirin as an J Am Coll Cardiol 2003; 41:961–965
antithrombotic therapeutic modality: the effect of low-dose 107 Wang T, Bhatt D, Topol E. Aspirin and clopidogrel resis-

tance: an emerging clinical entity. Eur Heart J 2006; 27:647– 127 Wardlaw J, Keir S, Dennis M. The impact of delays in
654 computed tomography of the brain on the accuracy of diagnosis
108 Hennekens C, Schror K, Weisman S, et al. Terms and and subsequent management in patients with minor stroke.
conditions: semantic complexity and aspirin resistance. Cir- J Neurol Neurosurg Psychiatry 2003; 74:77– 81
culation 2004; 110:1706 –1708 128 Keir S, Wardlaw J, Warlow C. Stroke epidemiology studies
109 Cipollone F, Ciabattoni G, Patrignani P, et al. Oxidant stress have underestimated the frequency of intracerebral hemor-
and aspirin-insensitive thromboxane biosynthesis in severe rhage: a systematic review of imaging in epidemiological
unstable angina. Circulation 2000; 102:1007–1013 studies. J Neurol 2002; 249:1226 –1231
110 Catella-Lawson F, Reilly M, Kapoor S, et al. Cyclooxygenase 129 Special report: preliminary report of the stroke prevention in
inhibitors and the antiplatelet effects of aspirin. N Engl atrial fibrillation study. N Engl J Med 1990; 322:863– 868
J Med 2001; 345:1809 –1817 130 Stroke Prevention in Atrial Fibrillation Study. Final results.
111 Capone M, Sciulli M, Tacconelli S, et al. Pharmacodynamic Circulation 1991; 84:527–539
interaction of naproxen with low-dose aspirin in healthy 131 Boston Area Anticoagulation Trial for Atrial Fibrillation
subjects. J Am Coll Cardiol 2005; 45:1295–1301 Investigators. The effect of low-dose warfarin on the risk of
112 Renda G, Tacconelli S, Capone M, et al. Celecoxib, ibupro- stroke in patients with nonrheumatic atrial fibrillation.
fen, and the antiplatelet effect of aspirin in patients with N Engl J Med 1990; 323:1505–1511
osteoarthritis and ischemic heart disease. Clin Pharmacol 132 Petersen P, Boysan G, Godtfredsen J, et al. Placebo-
Ther 2006; 80:264 –274 controlled, randomized trial of warfarin and aspirin for
113 Mac Donald T, Wei L. Effect of ibuprofen on cardioprotec- prevention of thromboembolic complications in chronic
tive effect of aspirin. Lancet 2003; 361:573–574 atrial fibrillation: the Copenhagen AFASAK Study. Lancet
114 Kurth T, Glynn R, Walker A, et al. Inhibition of clinical benefits 1989; 1:175–179
of aspirin on first myocardial infarction by nonsteroidal antiin- 133 EAFT (European Atrial Fibrillation Trial) Study Group.
flammatory drugs. Circulation 2003; 108:1191–1195 Secondary prevention in nonrheumatic atrial fibrillation
115 Garcı́a Rodrı́guez L, Varas-Lorenzo C, Maguire A, et al. after transient ischemic attack or minor stroke. Lancet 1993;
Nonsteroidal antiinflammatory drugs and the risk of myo- 342:1255–1262
cardial infarction in the general population. Circulation 134 Albers G. Atrial fibrillation and stroke: three new studies,
2004; 109:3000 –3006 three remaining questions. Arch Intern Med 1994; 154:
116 Eikelboom J, Hirsh J, Weitz J, et al. Aspirin-resistant 1443–1448
thromboxane biosynthesis and the risk of myocardial infarc- 135 Stroke Prevention in Atrial Fibrillation Investigators. Adjusted-
tion, stroke, or cardiovascular death in patients at high risk dose warfarin versus low-intensity, fixed-dose warfarin plus
for cardiovascular events. Circulation 2002; 105:1650 –1655 aspirin for high-risk patients with atrial fibrillation: Stroke
117 Patrono C, Rocca B. Drug insight: aspirin resistance; fact or Prevention in Atrial Fibrillation III randomized clinical trial.
fashion? Nat Clin Pract Cardiovasc Med 2007; 4:42–50 Lancet 1996; 348:633– 638
118 Patrono C, Bachmann F, Baigent C, et al. Expert Consensus 136 Hart R, Halperin J. Atrial fibrillation and stroke: revisiting
Document on the Use of Antiplatelet Agents: the task force the dilemmas. Stroke 1994; 25:1337–1341
on the use of antiplatelet agents in patients with atheroscle- 137 Stroke Prevention in Atrial Fibrillation Investigators. Pa-
rotic cardiovascular disease of the European Society of tients with nonvalvular atrial fibrillation at low risk of stroke
Cardiology. Eur Heart J 2004; 25:166 –181 during treatment with aspirin: Stroke Prevention in Atrial
119 Michelson A, Cattaneo M, Eikelboom J, et al. Aspirin Fibrillation III Study. JAMA 1998; 279:1273–1277
resistance: position paper of the Working Group on Aspirin 138 Pulmonary Embolism Prevention (PEP) Trial Collaborative
Resistance. J Thromb Haemost 2005; 3:1309 –1311 Group. Prevention of pulmonary embolism and deep vein
120 van Kooten F, Ciabattoni G, Patrono C, et al. Evidence for thrombosis with low dose aspirin: Pulmonary Embolism
episodic platelet activation in acute ischemic stroke. Stroke Prevention (PEP) trial. Lancet 2000; 355:1295–1302
1994; 25:278 –281 139 Antiplatelet Trialists’ Collaboration. Collaborative overview
121 Peto R, Gray R, Collins R, et al. Randomized trial of of randomized trials of antiplatelet therapy: III. Reduction
prophylactic daily aspirin in British male doctors. BMJ 1988; in venous thrombosis and pulmonary embolism by antiplate-
296:313–316 let prophylaxis among surgical and medical patients. BMJ
122 Hansson L, Zanchetti A, Carruthers S, et al. Effects of 1994; 308:235–243
intensive blood-pressure lowering and low-dose aspirin in 140 Mohr D, Silverstein M, Murtaugh P, et al. Prophylactic
patients with hypertension: principal results of the Hyper- agents for venous thrombosis in elective hip surgery. Arch
tension Optimal Treatment (HOT) randomized trial. Lancet Intern Med 1993; 253:2221–2228
1998; 351:1755–1762 141 Gallus A, Salzman E, Hirsh J, et al. Prevention of venous
123 Medical Research Council’s General Practice Research thrombo-embolism. Philadelphia, PA: JB Lippincott, 1994
Framework. Thrombosis prevention trial: randomized trial 142 Cruickshank M, Levine M, Hirsh J, et al. An evaluation of
of low-intensity oral anticoagulation with warfarin and low- impedance plethysmography and I-125-fibrinogen leg scan-
dose aspirin in the primary prevention of ischemic heart ning in patients following hip surgery. Thromb Haemost
disease in men at increased risk. Lancet 1998; 351:233–241 1989; 62:830 – 834
124 Collaborative Group of the Primary Prevention Project. 143 Powers P, Gent M, Jay R, et al. A randomized trial of less
Low-dose aspirin and vitamin E in people at cardiovascular intensive post-operative warfarin and aspirin therapy in the
risk: a randomized trial in general practice. Lancet 2001; prevention of venous thrombo-embolism after surgery for
357:89 –95 fractured hip. Arch Intern Med 1989; 149:771–774
125 Ridker P, Cook N, Lee I, et al. A randomized trial of 144 Graor R, Steward J, Lotke P, et al. RD Heparin (ardeparin
low-dose aspirin in the primary prevention of cardiovascular sodium) vs aspirin to prevent deep venous thrombosis after
disease in women. N Engl J Med 2005; 352:1293–1304 hip or knee replacement surgery [abstract]. Chest 1992;
126 Dennis M, Bamford J, Molyneux A, et al. Rapid resolution of 102(suppl):118S
signs of primary intracerebral hemorrhage in computed 145 Gent M, Hirsh J, Ginsberg J, et al. The low molecular weight
tomograms of the brain. BMJ 1987; 295:379 –381 heparinoid orgaran is more effective than aspirin in the

prevention of venous thromboembolism following surgery aspirin as plain and enteric-coated formulations. BMC Clin
for hip fracture. Circulation 1996; 93:80 – 84 Pharmacol 2001; 1:1
146 Imperiale T, Stollenwerk-Petrulis A. A meta-analysis of low 164 Yeomans N, Tulassay Z, Juhàsz L, et al. A comparison of
dose aspirin for the prevention of pregnancy-induced hyper- omeprazole with ranitidine for ulcers associated with non-
tensive disease. JAMA 1991; 266:260 –264 steroidal anti-inflammatory drugs. N Engl J Med 1998;
147 Uzan S, Beaufils M, Breart G, et al. Prevention of fetal 388:719 –726
growth retardation with low-dose aspirin: findings of the 165 Hawkey C, Karrasch J, Szczepañski L, et al. Omeprazole
EPREDA trial. Lancet 1991; 337:1427–1431 compared with misoprostol for ulcers associated with non-
148 Italian Study of Aspirin in Pregnancy. Low-dose aspirin in steroidal anti-inflammatory drugs. N Engl J Med 1998;
prevention and treatment of intrauterine growth retardation 338:727–734
and pregnancy-induced hypertension. Lancet 1993; 341: 166 Chan F, Chung S, Suen B, et al. Preventing recurrent upper
396 – 400 gastrointestinal bleeding in patients with Helicobacter pylori
infection who are taking low-dose aspirin or naproxen.
149 Sibai B, Caritis S, Thom E, et al. Prevention of preeclampsia
N Engl J Med 2001; 344:967–973
with low-dose aspirin in healthy, nulliparous pregnant
167 Chan F, Ching J, Hung L, et al. Clopidogrel versus aspirin
women. N Engl J Med 1993; 329:1213–1218
and esomeprazole to prevent recurrent ulcer bleeding.
150 CLASP (Collaborative Low-dose Aspirin Study in Preg- N Engl J Med 2005; 352:238 –244
nancy) Collaborative Group. CLASP: a randomized trial of 168 Lai K, Chu K, Hui W, et al. Esomeprazole with aspirin
low-dose aspirin for the prevention and treatment of pre- versus clopidogrel for prevention of recurrent gastrointesti-
eclampsia among 9,364 pregnant women. Lancet 1994; nal ulcer complications. Clin Gastroenterol Hepatol 2006;
343:619 – 629 4:860 – 865
151 ECPPA (Estudo Colaborativo para Prevencào da Pré- 169 Braunwald E, Antman E, Beasley J, et al. ACC/AHA
eclampsia com Aspirina) Collaborative Group. ECPPA: ran- guideline update for the management of patients with
domized trial of low dose aspirin for the prevention of unstable angina and non-ST-segment elevation myocardial
maternal and fetal complications in high-risk pregnant infarction-2002: summary article; a report of the American
women. Br J Obstet Gynaecol 1996; 103:39 – 47 College of Cardiology/American Heart Association Task
152 Rotchell Y, Cruickshank J, Phillips-Gay M, et al. Barbados Force on Practice Guidelines (Committee on the Manage-
Low dose Aspirin Study in Pregnancy (BLASP): a random- ment of Patients With Unstable Angina). Circulation 2002;
ized trial for the prevention of pre-eclampsia and its com- 106:1893–1900
plications. Br J Obstet Gynaecol 1998; 105:286 –292 170 Anderson JL, Adams CD, Antman EM, et al. ACC/AHA
153 Golding J. A randomized trial of low dose aspirin for 2007 guidelines for the management of patients with unsta-
primiparae in pregnancy: the Jamaica Low Dose Aspirin ble angina/non-ST-elevation myocardial infarction: a report
Study Group. Br J Obstet Gynaecol 1998; 105:293–299 of the American College of Cardiology/American Heart
154 Caritis S, Sibai B, Hauth J, et al. Low-dose aspirin to prevent Association Task Force on Practice Guidelines (Writing
preeclampsia in women at high risk. N Engl J Med 1998; Committee to Revise the 2002 Guidelines for the Manage-
338:701–705 ment of Patients With Unstable Angina/Non-ST-Elevation
155 Duley L, Henderson-Smart D, Knight M, et al. Antiplatelet Myocardial Infarction). Circulation 2007; 116:e148 – e304
drugs for prevention of pre-eclampsia and its consequences: 171 Iso H, Hennekens C, Stampfer M, et al. Prospective study of
systematic review. BMJ 2001; 322:329 –333 aspirin use and risk of stroke in women. Stroke 1999;
156 Roderick P, Wilkes H, Meade T. The gastrointestinal toxicity 30:1764 –1771
of aspirin: an overview of randomized controlled trials. Br J 172 Menè P, Pugliese F, Patrono C. The effects of nonsteroidal
Clin Pharmacol 1993; 35:219 –226 anti-inflammatory drugs on human hypertensive vascular
157 Garcı́a Rodrı́guez L, Cattaruzzi C, Troncon M, et al. Risk of disease. Semin Nephrol 1995; 15:244 –252
hospitalization for upper gastrointestinal tract bleeding associ- 173 Pierucci A, Simonetti B, Pecci G, et al. Improvement of
ated with ketorolac, other nonsteroidal anti-inflammatory renal function with selective thromboxane antagonism in
drugs, calcium antagonists, and other antihypertensive drugs. lupus nephritis. N Engl J Med 1989; 320:421– 425
Arch Intern Med 1998; 158:33–39 174 Al-Khadra A, Salem D, Rand W, et al. Antiplatelet agents
158 Weil J, Colin-Jones D, Langman M, et al. Prophylactic and survival: a cohort analysis from the studies of left
aspirin and risk of peptic ulcer bleeding. BMJ 1995; 310: ventricular dysfunction (SOLVD) trial. J Am Coll Cardiol
827– 830 1998; 31:419 – 425
159 Garcı́a Rodrı́guez L, Hernández-Dı́az S, de Abajo F. Asso- 175 Latini R, Tognoni G, Maggioni A, et al. Clinical effects of
ciation between aspirin and upper gastrointestinal compli- early ACE inhibitor treatment for acute myocardial infarc-
cations: systematic review of epidemiologic studies. Br J Clin tion are similar in the presence and absence of aspirin:
Pharmacol 2001; 52:563–571 meta-analysis of individual data from nearly 96,000 random-
160 Wolfe M, Lichtenstein D, Singh G. Gastrointestinal toxicity ized patients. J Am Coll Cardiol 2000; 35:1801–1807
of nonsteroidal antiinflammatory drug. N Engl J Med 1999; 176 Zanchetti A, Hansson L, Leonetti G, et al. Low-dose aspirin
340:1888 –1899 does not interfere with the blood pressure-lowering effects
161 Lanas A, Perez-Aisa M, Feu F, et al. A nationwide study of of antihypertensive therapy. J Hypertens 2002; 20:1015–
mortality associated with hospital admission due to severe 1022
gastrointestinal events and those associated with nonsteroi- 177 Teo K, Yusuf S, Pfeffer M, et al. Effects of long-term
dal antiinflammatory drug use. Am J Gastroenterol 2005; treatment with angiotensin-converting-enzyme inhibitors in
100:1685–1693 the presence or absence of aspirin: a systematic review.
162 Kelly J, Kaufman D, Jurgelon J, et al. Risk of aspirin-associated Lancet 2002; 360:1037–1043
major upper-gastrointestinal bleeding with enteric-coated or 178 Burke A, Smyth E, FitzGerald G. Analgesic-antipyretic and
buffered product. Lancet 1996; 348:1413–1416 anti-inflammatory agents and drugs employed in the treat-
163 De Abajo F, Garcı́a Rodrı́guez L. Risk of upper gastrointes- ment of gout. In: Goodman & Gilman’s the pharmacological
tinal bleeding and perforation associated with low-dose basis of therapeutics. New York, NY: McGraw-Hill, 2006

179 Capone M, Tacconelli S, Sciulli M, et al. Clinical pharma- 196 Solomon S, McMurray J, Pfeffer M, et al. Cardiovascular
cology of platelet, monocyte and vascular cyclooxygenase risk associated with celecoxib in a clinical trial for colorectal
inhibition by naproxen and low-dose aspirin in healthy adenoma prevention. N Engl J Med 2005; 352:1071–1080
subjects. Circulation 2004; 109:1468 –1471 197 Bresalier R, Sandler R, Quan H, et al. Cardiovascular events
180 Pedersen A, FitzGerald G. Cyclooxygenase inhibition, plate- associated with rofecoxib in a colorectal adenoma chemo-
let function and drug metabolite formation during chronic prevention trial. N Engl J Med 2005; 352:1092–1102
administration of sulfinpyrazone in man. Clin Pharmacol 198 Fitzgerald G. Coxibs and cardiovascular disease. N Engl
Ther 1985; 37:36 – 42 J Med 2004; 351:1709 –1711
181 Rebuzzi A, Natale A, Bianchi C, et al. Effects of indobufen 199 Cannon C, Curtis S, FitzGerald G, et al. Cardiovascular
on platelet thromboxane B2 production in patients with outcomes with etoricoxib and diclofenac in patients with
myocardial infarction. Eur J Clin Pharmacol 1990; 39:99 – osteoarthritis and rheumatoid arthritis: results from the
100 multinational etoricoxib and diclofenac arthritis long-term
182 Davı̀ G, Patrono C, Catalano I, et al. Inhibition of throm- (MEDAL) Program. Lancet 2006; 368:1771–1781
boxane biosynthesis and platelet function by indobufen in 200 Born G, Patrono C. Antiplatelet drug. Br J Pharmacol 2006;
type II diabetes mellitus. Arterioscler Thromb 1993; 13: 147(suppl):S241–S251
1346 –1349 201 Weyrich A, Denis M, Kuhlmann-Eyre J, et al. Dipyridamole
183 Rajah S, Rees M, Walker D, et al. Effects of antiplatelet selectively inhibits inflammatory gene expression in platelet-
therapy with indobufen or aspirin-dipyridamole on graft monocyte aggregates. Circulation 2005; 111:633– 642
patency 1 year after coronary artery bypass grafting. J Thorac 202 Müller T, Su C, Weisenberger H, et al. Dipyridamole alone
Cardiovasc Surg 1994; 107:1146 –1153 or combined with low-dose acetylsalicylic acid inhibits plate-
184 SINBA Group. Indobufen versus aspirin plus dipyridamole let aggregation in human whole blood ex vivo. Br J Clin
after coronary artery surgery. Coron Artery Dis 1991; 2:897– Pharmacol 1990; 30:179 –186
906 203 De Schryver E, Algra A, van Gijn J. Dipyridamole for
185 Fornaro G, Rossi P, Mantica P, et al. Indobufen in the preventing stroke and other vascular events in patients with
prevention of thrombo-embolic complications in patients vascular disease. Cochrane Database Syst Rev 2006;
with heart disease. Circulation 1993; 87:162–164 2:CD001820
186 Morocutti C, Amabile G, Fattapposta F, et al. Indobufen 204 Diener H, Darius H, Bertrand-Hardy J, et al. European
versus warfarin in the secondary prevention of major vascu- Stroke Prevention Study 2: cardiac safety in the European
lar events in nonrheumatic atrial fibrillation. Stroke 1997; Stroke Prevention Study 2 (ESPS2). Int J Clin Pract 2001;
28:1015–1021 55:162–163
187 Bergamasco B, Benna P, Carolei A, et al. A randomized trial 205 ESPRIT Study Group. Aspirin plus dipyridamole versus
comparing ticlopidine hydrochloride with indobufen for the aspirin alone after cerebral ischaemia of arterial origin
prevention of stroke in high-risk patients (TISS Study): (ESPRIT): randomized controlled trial. Lancet 2006; 367:
Ticlopidine Indobufen Stroke Study. Funct Neurol 1997; 1665–1673
12:33– 43 206 Savi P, Herbert J. Clopidogrel and ticlopidine: p2Y12 aden-
188 Brochier M. Evaluation of flurbiprofen for prevention of osine diphosphate-receptor antagonists for the prevention of
reinfarction and reocclusion after successful thrombolysis or atherothrombosis. Semin Thromb Hemost 2005; 31:174 –
angioplasty in acute myocardial infarction: the Flurbiprofen 183
French Trial. Eur Heart J 1993; 14:951–957 207 Savi P, Pereillo J, Uzabiaga M, et al. Identification and
189 Ramis J, Torrent J, Mis R, et al. Pharmacokinetics of triflusal biological activity of the active metabolite of clopidogrel.
after single and repeated doses in man. Int J Clin Pharmacol Thromb Haemost 2000; 84:891– 896
Ther Toxicol 1990; 28:344 –349 208 Pereillo J, Maftouh M, Andrieu A, et al. Structure and
190 Cruz-Fernandez J, Lopez-Bescos L, Garcı́a-Dorado D, et al. stereochemistry of the active metabolite of clopidogrel.
Randomized comparative trial of triflusal and aspirin follow- Drug Metab Dispos 2002; 30:1288 –1295
ing acute myocardial infarction. Eur Heart J 2000; 21:457– 209 Hollopeter G, Jantzen H, Vincent D, et al. Identification of
465 the platelet ADP receptor targeted by antithrombotic drugs.
191 Matı́as-Guiu J, Ferro J, Alvarez-Sabı́n J, et al. Comparison of Nature 2001; 409:202–207
triflusal and aspirin for prevention of vascular events in 210 Ding Z, Kim S, Dorsam R, et al. Inactivation of the human
patients after cerebral infarction: the TACIP Study; a ran- P2Y12 receptor by thiol reagents requires interaction with
domized, double blind, multicenter trial. Stroke 2003; 34: both extracellular cysteine residues Cys17 and Cys270.
840 – 848 Blood 2003; 101:3908 –3914
192 Silverstein F, Faich G, Goldstein J, et al. Gastrointestinal 211 Savi P, Zachayus J, Delesque-Touchard N, et al. The active
toxicity with celecoxib vs. nonsteroidal anti-inflammatory drugs metabolite of Clopidogrel disrupts P2Y12 receptor oli-
for osteoarthritis and rheumatoid arthritis: the CLASS study; a gomers and partitions them out of lipid rafts. Proc Natl Acad
randomized controlled trial. JAMA 2000; 284:1247–1255 Sci U S A 2006; 103:11069 –11074
193 Bombardier C, Laine L, Reicin A, et al. Comparison of 212 Cattaneo M, Zighetti M, Lombardi R, et al. Molecular bases
upper gastrointestinal toxicity of rofecoxib and naproxen in of defective signal transduction in the platelet P2Y12 recep-
patients with rheumatoid arthritis. N Engl J Med 2000; tor of a patient with congenital bleeding. Proc Natl Acad Sci
343:1520 –1528 U S A 2003; 100:1978 –1983
194 Schnitzer T, Burmester G, Mysler E, et al. Comparison of 213 Balsano F, Rizzon P, Violi F, et al. Antiplatelet treatment
lumiracoxib with naproxen and ibuprofen in the Thera- with ticlopidine in unstable angina: a controlled multicenter
peutic Arthritis Research and Gastrointestinal Event Trial clinical trial. Circulation 1990; 82:17–26
(TARGET), reduction in ulcer complications: randomized 214 Gent M, Blakely J, Easton J, et al. The Canadian American
controlled trial. Lancet 2004; 364:665– 674 Ticlopidine Study (CATS) in thromboembolic stroke. Lan-
195 Nussmeier N, Whelton A, Brown M, et al. Complications of cet 1989; 2:1215–1220
the COX-2 inhibitors parecoxib and valdecoxib after cardiac 215 Hass W, Easton J, Adams H, et al. A randomized trial
surgery. N Engl J Med 2005; 352:1081–1091 comparing ticlopidine hydrochloride with aspirin for the

prevention of stroke in high-risk patients. N Engl J Med bution of gene sequence variations of the hepatic cyto-
1989; 321:501–507 chrome P450 3A4 enzyme to variability in individual respon-
216 Scrutinio D, Cimminiello C, Marubini E, et al. Ticlopidine siveness to clopidogrel. Arterioscler Thromb Vasc Biol 2006;
versus aspirin after myocardial infarction (STAMI) trial. 26:1895–1900
J Am Coll Cardiol 2001; 37:1259 –1265 236 Savcic M, Hauert J, Bachmann F, et al. Clopidogrel loading
217 Arcan J, Blanchard J, Boissel J, et al. Multicenter double dose regimens: kinetic profile of pharmacodynamic response
blind study of ticlopidine in the treatment of intermittent in healthy subjects. Semin Thromb Hemost 1999; 25:15–19
claudication and the prevention of its complications. Angi- 237 Hochholzer W, Trenk D, Frundi D, et al. Time dependence
ology 1988; 39:802– 811 of platelet inhibition after a 600-mg loading dose of clopi-
218 Balsano F, Coccheri S, Libretti A, et al. Ticlopidine in the dogrel in a large, unselected cohort of candidates for
treatment of intermittent claudication: a 21-month double percutaneous coronary intervention. Circulation 2005; 111:
blind trial. J Lab Clin Med 1989; 114:84 –91 2560 –2564
219 Janzon L, Bergqvist D, Boberg J, et al. Prevention of 238 von Beckerath N, Taubert D, Pogatsa-Murray G, et al.
myocardial infarction and stroke in patients with intermit- Absorption, metabolization, and antiplatelet effects of 300-,
tent claudication: effects of ticlopidine; results from STIMS, 600-, and 900-mg loading doses of clopidogrel: results of the
the Swedish Ticlopidine Multicentre Study. J Intern Med ISAR-CHOICE (Intracoronary Stenting and Antithrom-
1990; 227:301–308 botic Regimen; Choose Between 3 High Oral Doses for
220 Limet R, David J, Magotteaux P, et al. Prevention of Immediate Clopidogrel Effect) Trial. Circulation 2005; 112:
aortocoronary bypass graft occlusion: beneficial effect of 2946 –2950
ticlopidine on early and late patency rate of venous coronary 239 Montalescot G, Sideris G, Meuleman C, et al. A randomized
bypass grafts; a double-blind study. J Thorac Cardiovasc comparison of high clopidogrel loading doses in patients
Surg 1987; 94:773–783 with non-ST-segment elevation acute coronary syndromes:
221 FitzGerald G. Ticlopidine in unstable angina: a more expen- the ALBION (Assessment of the Best Loading Dose of
sive aspirin? Circulation 1990; 82:296 –298 Clopidogrel to Blunt Platelet Activation, Inflammation and
222 Yeh S, Hsueh E, Wu H, et al. Ticlopidine-associated aplastic Ongoing Necrosis) trial. J Am Coll Cardiol 2006; 48:931–938
anemia: a case report and review of literature. Ann Hematol 240 Cuisset T, Frere C, Quilici J, et al. Benefit of a 600-mg
1998; 76:87–90 loading dose of clopidogrel on platelet reactivity and clinical
223 Bennett C, Davidson C, Raisch D, et al. Thrombotic outcomes in patients with non-ST-segment elevation acute
thrombocytopenic purpura associated with ticlopidine in the coronary syndrome undergoing coronary stenting. J Am Coll
setting of coronary artery stents and stroke prevention. Arch Cardiol 2006; 48:1339 –1345
Intern Med 1999; 159:2524 –2528 241 Jaremo P, Lindahl T, Fransson S, et al. Individual variations
224 Herbert J, Bernat A, Samama M, et al. The anti aggregating of platelet inhibition after loading doses of clopidogrel.
and antithrombotic activity of ticlopidine is potentiated by J Intern Med 2002; 252:233–238
aspirin in the rat. Thromb Haemost 1996; 76:94 –98 242 Gurbel P, Bliden K, Hiatt B, et al. Clopidogrel for coronary
225 Lecompte T, Lecrubier C, Bouloux C, et al. Antiplatelet stenting: response variability, drug resistance, and the effect
effects of the addition of acetylsalicylic acid 40 mg daily to of pretreatment platelet reactivity. Circulation 2003; 107:
ticlopidine in human healthy volunteers. Clin Appl Thromb 2908 –2913
Hemost 1997; 3:245–250 243 Muller I, Besta F, Schulz C, et al. Prevalence of clopidogrel
226 Leon M, Baim D, Popma J, et al. A clinical trial comparing non-responders among patients with stable angina pectoris
three antithrombotic-drug regimens after coronary artery scheduled for elective coronary stent placement. Thromb
stenting. N Engl J Med 1998; 339:1665–1671 Haemost 2003; 89:783–787
227 More R, Chauhan A. Antiplatelet rather than anticoagulant 244 Lau W, Waskell L, Watkins P, et al. Atorvastatin reduces the
therapy with coronary stenting. Lancet 1997; 349:146 –147 ability of clopidogrel to inhibit platelet aggregation: a new
228 Bhatt D, Bertrand M, Berger P, et al. Meta-analysis of drug-drug interaction. Circulation 2003; 107:32–37
randomized and registry comparisons of ticlopidine with 245 Neubauer H, Gunesdogan B, Hanefeld C, et al. Lipophilic
clopidogrel after stenting. J Am Coll Cardiol 2002; 39:9 –14 statins interfere with the inhibitory effects of clopidogrel on
229 Steinhubl S, Tan W, Foody J, et al. Incidence and clinical platelet function: a flow cytometry study. Eur Heart J 2003;
course of thrombotic thrombocytopenic purpura due to 24:1744 –1749
ticlopidine following coronary stenting: EPISTENT Investi- 246 Wilkinson G. Drug metabolism and variability among pa-
gators; Evaluation of Platelet IIb/IIIa Inhibitor for Stenting. tients in drug response. N Engl J Med 2005; 352:2211–2221
JAMA 1999; 281:806 – 810 247 Wienbergen H, Gitt A, Schiele R, et al. Comparison of
230 Herbert J, Frehel D, Vallee E, et al. Clopidogrel, a novel clinical benefits of clopidogrel therapy in patients with acute
antiplatelet and antithrombotic agent. Cardiovasc Drug Rev coronary syndromes taking atorvastatin versus other statin
1993; 11:180 –198 therapies. Am J Cardiol 2003; 92:285–288
231 CAPRIE Steering Committee. A randomized, blinded, trial 248 Saw J, Steinhubl S, Berger P, et al. Lack of adverse
of clopidogrel versus aspirin in patients at risk of ischemic clopidogrel–atorvastatin clinical interaction from secondary
events (CAPRIE). Lancet 1996; 348:1329 –1339 analysis of a randomized, placebo-controlled clopidogrel
232 Taubert D, Kastrati A, Harlfinger S, et al. Pharmacokinetics trial. Circulation 2003; 108:921–924
of clopidogrel after administration of a high loading dose. 249 Mukherjee D, Kline-Rogers E, Fang J, et al. Lack of
Thromb Haemost 2004; 92:311–316 clopidogrel-CYP3A4 statin interaction in patients with acute
233 Clarke T, Waskell L. The metabolism of clopidogrel is coronary syndrome. Heart 2005; 91:23–26
catalyzed by human cytochrome P450 3A and is inhibited by 250 Lim M, Spencer F, Gore J, et al. Impact of combined
atorvastatin. Drug Metab Dispos 2003; 31:53–59 pharmacologic treatment with clopidogrel and a statin on
234 Lau W, Gurbel P, Watkins P, et al. Contribution of hepatic outcomes of patients with non-ST-segment elevation acute
cytochrome P450 3A4 metabolic activity to the phenomenon coronary syndromes: perspectives from a large multinational
of clopidogrel resistance. Circulation 2004; 109:166 –171 registry. Eur Heart J 2005; 26:1063–1069
235 Angiolillo D, Fernandez-Ortiz A, Bernardo E, et al. Contri- 251 Gurbel P, Tantry U. Drug insight: clopidogrel nonrespon-

siveness. Nat Clin Pract Cardiovasc Med 2006; 3:387–395 prolonged platelet inhibition with gradual recovery from GP
252 Bennett C, Connors J, Carwile J, et al. Thrombotic throm- IIb/IIIa receptor blockade. Circulation 1998; 97:1680 –1688
bocytopenic purpura associated with clopidogrel. N Engl 270 EPIC Investigators. Use of a monoclonal antibody directed
J Med 2000; 342:1773–1777 against the platelet glycoprotein IIb/IIIa receptor in high-
253 Clopidogrel in Unstable Angina to Prevent Recurrent risk coronary angioplasty. N Engl J Med 1994; 330:956 –961
Events Trial Investigators. Effects of clopidogrel in addition 271 Peter K, Kohler B, Straub A, et al. Flow cytometric moni-
to aspirin in patients with acute coronary syndromes without toring of glycoprotein IIb/IIIa blockade and platelet func-
ST-segment elevation. N Engl J Med 2001; 345:494 –502 tion in patients with acute myocardial infarction receiving
254 Steinhubl S, Berger P, Tift Mann J III, et al. Early and reteplase, abciximab, and ticlopidine: continuous platelet
sustained dual oral antiplatelet therapy following percutane- inhibition by the combination of abciximab and ticlopidine.
ous coronary intervention. JAMA 2002; 288:2411–2420 Circulation 2000; 102:1490 –1496
255 Diener H, Bogousslavsky J, Brass L, et al. Aspirin and 272 Nakada M, Sassoli P, Tam S, et al. Abciximab pharmacody-
clopidogrel compared with clopidogrel alone after recent
namics are unaffected by antecedent therapy with other
ischemic stroke or transient ischemic attack in high-risk
GPIIb/IIIa antagonists in non-human primates. Thromb
patients (MATCH): randomized, double blind, placebo-
Thrombolysis 2002; 14:15–24
controlled trial. Lancet 2004; 364:331–337
256 Sabatine M, Cannon C, Gibson C, et al. Addition of 273 Quinn M, Deering A, Stewart M, et al. Quantifying GPIIb/
clopidogrel to aspirin and fibrinolytic therapy for myocardial IIIa receptor binding using 2 monoclonal antibodies: dis-
infarction with ST-segment elevation. N Engl J Med 2005; criminating abciximab and small molecular weight antago-
352:1179 –1189 nists. Circulation 1999; 99:2231–2238
257 Chen Z, Jiang L, Chen Y, et al. Addition of clopidogrel to 274 Lev E, Osende J, Richard M, et al. Administration of abciximab
aspirin in 45,852 patients with acute myocardial infarction: to patients receiving tirofiban or eptifibatide: effect on platelet
randomized placebo-controlled trial. Lancet 2005; 366: function. J Am Coll Cardiol 2001; 37:847– 855
1607–1621 275 EPILOG Investigators. Platelet glycoprotein IIb/IIIa recep-
258 Bhatt D, Fox K, Hacke W, et al. Clopidogrel and aspirin tor blockade and low-dose heparin during percutaneous
versus aspirin alone for the prevention of atherothrombotic coronary revascularization. N Engl J Med 1997; 336:1689 –
events. N Engl J Med 2006; 354:1706 –1717 1696
259 ACTIVE Writing Group on behalf of the ACTIVE Investi- 276 Kereiakes DJ, Essel JH, Abbottsmith CW, et al. Abciximab-
gators. Clopidogrel plus aspirin versus oral anticoagulation associated profound thromborytopenia: therapy with immu-
for atrial fibrillation in the Atrial fibrillation Clopidogrel noglobulin and platelet transfusion. Am J Cardiol 1996;
Trial with Irbesartan for prevention of Vascular Events 78:1161–1163
(ACTIVE W): a randomized controlled trial. Lancet 2006; 277 Berkowitz SD, Harrington RA, Rund MM, et al. Acute
367:1903–1912 profound thrombocytopenia following c7E3 Fab (abcix-
260 Coller B. Platelet GPIIb/IIIa antagonists: the first anti- imab) therapy. Circulation 1997; 95:809 – 813
integrin receptor therapeutics. J Clin Invest 1997; 99:1467– 278 Madan M, Berkowitz S. Understanding thrombocytopenia
1471 and antigenicity with glycoprotein IIb-IIIa inhibitors. Am
261 Coller B. Binding of abciximab to alpha V beta 3 and Heart J 1999; 138:317–326
activated alpha M beta 2 receptors: with a review of 279 Curtis B, Swyerr J, Divgi A, et al. Thrombocytopenia after
platelet-leukocyte interactions. Thromb Haemost 1999; 82: second exposure to abciximab is caused by antibodies that
326 –336 recognize abciximab-coated platelets. Blood 2002; 99:2054 –
262 Schwarz M, Nordt T, Bode C, et al. The GPIIb/IIIa 2059
inhibitor abciximab (c7E3) inhibits the binding of various 280 Kaluski E, Leitman M, Khiger I, et al. Delayed thrombocy-
ligands to the leukocyte integrin Mac-1 (CD11b/CD18, topenia following abciximab therapy. Int J Cardiovasc Inter-
alphasMbeta2). Thromb Res 2002; 107:121–128 vent 2001; 4:151–155
263 Anderson K, Weisman H, Coller B. Platelet glycoprotein 281 Sharma S, Bhambi B, Nyitray W, et al. Delayed profound
IIb/IIIa receptor inhibitors. Philadelphia, PA: WB Saunders, thrombocytopenia presenting 7 days after use of abciximab
1999 (ReoPro). J Cardiovasc Pharmacol Ther 2002; 7:21–24
264 Topol E, Byzova T, Plow E. Platelet GPIIb/IIIa blockers. 282 Tcheng J, Kereiakes D, Lincoff A, et al. Abciximab read-
Lancet 1999; 353:227–231 ministration: results of the ReoPro Readministration Regis-
265 Scarborough R, Kleiman N, Phillips D. Platelet glycoprotein try. Circulation 2001; 104:870 – 875
IIb/IIIa antagonists: what are the relevant issues concerning 283 Madan M, Kereiakes D, Hermiller J, et al. Efficacy of
their pharmacology and clinical use? Circulation 1999; 100: abciximab readministration in coronary intervention. Am J
437– 444 Cardiol 2000; 85:435– 440
266 Nurden A, Poujol C, Durrieu-Jais C, et al. Platelet glycop- 284 Best P, Lennon R, Gersh B, et al. Safety of abciximab in
rotein IIb/IIIa inhibitors: basic and clinical aspects. Arterio- patients with chronic renal insufficiency who are undergoing
scler Thromb Vasc Biol 1999; 19:2835–2840 percutaneous coronary interventions. Am Heart J 2003;
267 Coller B, Peerschke E, Scudder L, et al. A murine mono- 146:345–350
clonal antibody that completely blocks the binding of fibrin- 285 Reverter J, Bèguin S, Kessels H, et al. Inhibition of platelet-
ogen to platelets produces a thrombasthenic-like state in mediated, tissue factor-induced thrombin generation by the
normal platelets and binds to glycoproteins IIb and/or IIIa. mouse/human chimeric 7E3 antibody: potential implications
J Clin Invest 1983; 72:325–338 for effect of c7E3 Fab treatment on acute thrombosis and
268 Tcheng J, Ellis S, George B, et al. Pharmacodynamics of ‘clinical restenosis.’ J Clin Invest 1996; 98:863– 874
chimeric glycoprotein IIb/IIIa integrin antiplatelet antibody 286 Hartman G, Egbertson M, Halczenko W, et al. Non-peptide
Fab 7E3 in high-risk coronary angioplasty. Circulation 1994; fibrinogen receptor antagonists: I. Discovery and design of
90:1757–1764 exosite inhibitors. J Med Chem 1992; 35:4640 – 4642
269 Mascelli M, Lance E, Damaraju L, et al. Pharmacodynamic 287 Egbertson M, Chang C, Duggan M, et al. Non-peptide
profile of a short-term abciximab treatment demonstrates fibrinogen receptor antagonists: II. Optimization of a ty-

rosine template as a mimic for Arg-Gly-Asp. J Med Chem receptor-directed platelet inhibition. Am Heart J 2001;
1994; 37:2537–2551 142:204 –210
288 Vickers S, Theoharides A, Arison B, et al. In vitro and in vivo 304 Tcheng J, Harrington R, Kottke Marchant K, et al. Multi-
studies on the metabolism of tirofiban. Drug Metab Dispos center, randomized, double-blind, placebo-controlled trial
1999; 27:1360 –1366 of the platelet integrin glycoprotein IIb/IIIa blocker integre-
289 Barrett J, Murphy G, Peerlinck K, et al. Pharmacokinetics lin in elective coronary intervention. Circulation 1995; 91:
and pharmacodynamics of MK-383, a selective non-peptide 2151–2157
platelet glycoprotein-IIb/IIIa receptor antagonist, in healthy 305 IMPACT-II Investigators. Randomized placebo-controlled
men. Clin Pharmacol Ther 1994; 56:377–388 trial of effect of eptifibatide on complications of percutane-
290 Peerlinck K, De Lepeleire I, Goldberg M, et al. MK-383 ous coronary intervention: IMPACT-I. Lancet 1997; 349:
(L-700,462), a selective nonpeptide platelet glycoprotein 1422–1428
IIb/IIIa antagonist, is active in man. Circulation 1993; 306 Harrington R, Kleiman N, Kottke Marchant K, et al.
Immediate and reversible platelet inhibition after intrave-
88:1512–1517
nous administration of a peptide glycoprotein IIb/IIIa inhib-
291 Kereiakes D, Kleiman N, Ambrose J, et al. Randomized,
itor during percutaneous intervention. Am J Cardiol 1995;
double-blind, placebo-controlled dose-ranging study of tiro-
76:1222–1227
fiban (MK-383) platelet IIb/IIIa blockade in high-risk pa- 307 Tardiff B, Jennings L, Harrington R, et al. Pharmacody-
tients undergoing coronary angioplasty. J Am Coll Cardiol namics and pharmacokinetics of eptifibatide in patients
1996; 27:536 –542 with acute coronary syndromes: prospective analysis from
292 Smith B, Gandhi P. Pharmacokinetics and pharmacodynam- PURSUIT. Circulation 2001; 104:399 – 405
ics of low-molecular-weight heparins and glycoprotein IIb/ 308 PURSUIT Trial Investigators. Inhibition of platelet glyco-
IIIa receptor antagonists in renal failure. J Thromb Throm- protein IIb/IIIa with eptifibatide in patients with acute
bolysis 2001; 11:39 – 48 coronary syndromes. N Engl J Med 1998; 339:436 – 443
293 Januzzi JJ, Snapinn S, DiBattiste P, et al. Benefits and safety 309 ESPRIT Investigators. Enhanced Suppression of the Plate-
of tirofiban among acute coronary syndrome patients with let IIb/IIIa Receptor with Integrilin Therapy: novel dosing
mild to moderate renal insufficiency: results from the Plate- regimen of eptifibatide in planned coronary stent implanta-
let Receptor Inhibition in Ischemic Syndrome Management tion (ESPRIT); a randomized, placebo-controlled trial. Lan-
in Patients Limited by Unstable Signs and Symptoms cet 2000; 356:2037–2044
(PRISM-PLUS) trial. Circulation 2002; 105:2361–2366 310 O’Shea J, Hafley G, Greenberg S, et al. Platelet glycoprotein
294 Bednar B, Cook J, Holahan M, et al. Fibrinogen receptor IIb/IIIa integrin blockade with eptifibatide in coronary stent
antagonist-induced thrombocytopenia in chimpanzee and intervention: the ESPRIT trial; a randomized controlled
rhesus monkey associated with preexisting drug-dependent trial. JAMA 2001; 285:2468 –2473
antibodies to platelet glycoprotein IIb/IIIa. Blood 1999; 311 Rezkalla S, Hayes J, Curtis B, et al. Eptifibatide-induced
94:587–599 acute profound thrombocytopenia presenting as refrac-
295 Bougie D, Wilker P, Wuitschick E, et al. Acute thrombocy- tory hypotension. Catheter Cardiovasc Interv 2003; 58:
topenia after treatment with tirofiban or eptifibatide is 76 –79
associated with antibodies specific for ligand-occupied 312 RESTORE Investigators. Effects of platelet glycoprotein
GPIIb/IIIa. Blood 2002; 100:2071–2076 IIb/IIIa blockade with tirofiban on adverse cardiac events in
296 Scarborough R, Naughton M, Teng W, et al. Design of patients with unstable angina or acute myocardial infarction
potent and specific integrin antagonists: peptide antagonists undergoing coronary angioplasty. Circulation 1997; 96:
with high specificity for glycoprotein IIb–IIIa. J Biol Chem 1445–1453
1993; 268:1066 –1073 313 CAPTURE Investigators. Randomized placebo-controlled
297 Lele M, Sajid M, Wajih N, et al. Eptifibatide and 7E3, but trial of abciximab before and during coronary intervention in
not tirofiban, inhibit alpha (v) beta (3) integrin-mediated refractory unstable angina: the CAPTURE study. Lancet
binding of smooth muscle cells to thrombospondin and 1997; 349:1429 –1435
prothrombin. Circulation 2001; 104:582–587 314 EPISTENT Investigators. Randomized placebo-controlled
298 Phillips D, Teng W, Arfsten A, et al. Effect of Ca2⫹ on GP and balloon-angioplasty-controlled trial to assess safety of
IIb–IIIa interactions with integrilin: enhanced GP IIb–IIIa coronary stenting with use of platelet glycoprotein-IIb/IIIa
binding and inhibition of platelet aggregation by reductions blockade. Lancet 1998; 352:87–92
in the concentration of ionized calcium in plasma anticoag- 315 Topol E, Lincoff A, Kereiakes D, et al. Multi-year follow-up of
ulated with citrate. Circulation 1997; 96:1488 –1494 abciximab therapy in three randomized, placebo-controlled trials
299 Alton K, Kosoglou T, Baker S, et al. Disposition of 14C- of percutaneous coronary revascularization. Am J Med 2002;
eptifibatide after intravenous administration to healthy men. 113:1– 6
Clin Ther 1998; 20:307–323 316 Stone G, Grines C, Cox D, et al. Comparison of angioplasty
300 Reddan D, O’Shea J, Sarembock I, et al. Treatment effects with stenting, with or without abciximab, in acute myocar-
of eptifibatide in planned coronary stent implantation in dial infarction. N Engl J Med 2002; 346:957–966
patients with chronic kidney disease (ESPRIT Trial). Am J 317 Kastrati A, Mehilli J, Schuhlen H, et al. A clinical trial of
Cardiol 2003; 91:17–21 abciximab in elective percutaneous coronary intervention
301 Gilchrist I, O’Shea J, Kosoglou T, et al. Pharmacodynamics after pretreatment with clopidogrel. N Engl J Med 2004;
and pharmacokinetics of higher-dose, double-bolus eptifi- 350:232–238
batide in percutaneous coronary intervention. Circulation 318 Hausleiter J, Kastrati A, Mehilli J, et al. A randomized trial
2001; 104:406 – 411 comparing phosphorylcholine-coated stenting with balloon
302 Phillips D, Scarborough R. Clinical pharmacology of eptifi- angioplasty as well as abciximab with placebo for restenosis
batide. Am J Cardiol 1997; 80:11B–20B reduction in small coronary arteries. J Intern Med 2004;
303 Li Y, Spencer F, Becker R. Comparative efficacy of fibrin- 256:388 –397
ogen and platelet supplementation on the in vitro reversibil- 319 Mehilli J, Kastrati A, Schuhlen H, et al. Randomized clinical
ity of competitive glycoprotein IIb/IIIa (alphaIIb/beta3) trial of abciximab in diabetic patients undergoing elective

percutaneous coronary interventions after treatment with a 326 GUSTO IV-ACS Investigators. Effect of glycoprotein IIb/
high loading dose of clopidogrel. Circulation 2004; 110: IIIa receptor blocker abciximab on outcome in patients with
3627–3635 acute coronary syndromes without early coronary revascu-
320 Kastrati A, Mehilli J, Neumann F, et al. Abciximab in larization: the GUSTO IV-ACS randomized trial. Lancet
patients with acute coronary syndromes undergoing percu- 2001; 357:1915–1924
taneous coronary intervention after clopidogrel pretreat- 327 Boersma E, Harrington R, Moliterno D, et al. Platelet
ment: the ISAR-REACT 2 randomized trial. JAMA 2006; glycoprotein IIb/IIIa inhibitors in acute coronary syn-
295:1531–1538 dromes: a meta-analysis of all major randomized clinical
321 Topol E, Moliterno D, Herrmann H, et al. Comparison of trials. Lancet 2002; 359:189 –198
two platelet glycoprotein IIb/IIIa inhibitors, tirofiban and
328 Antman E, Giugliano R, Gibson C, et al. Abciximab
abciximab, for the prevention of ischemic events with
facilitates the rate and extent of thrombolysis: results of
percutaneous coronary revascularization. N Engl J Med
the thrombolysis in myocardial infarction (TIMI) 14 trial:
2001; 344:1888 –1894
322 Moliterno D, Yakubov S, DiBattiste P, et al. Outcomes at 6 the TIMI 14 Investigators. Circulation 1999; 99:2720 –
months for the direct comparison of tirofiban and abciximab 2732
during percutaneous coronary revascularization with stent 329 GUSTO V Investigators. Reperfusion therapy for acute
placement: the TARGET follow-up study. Lancet 2002; myocardial infarction with fibrinolytic therapy or combina-
360:355–360 tion reduced fibrinolytic therapy and platelet glycoprotein
323 PARAGON Investigators. Randomized trial of potent plate- IIb/IIIa inhibition: the GUSTO V randomized trial. Lancet
let IIb/IIIa antagonism, heparin, or both in patients with 2001; 357:1905–1914
unstable angina: the PARAGON study [abstract]. Circula- 330 Lincoff A, Califf R, Van de Werf F, et al. Mortality at 1 year
tion 1996; 94:I553 with combination platelet glycoprotein IIb/IIIa inhibition
324 Platelet Receptor Inhibition in Ischemic Syndrome Manage- and reduce-dose fibrinolytic therapy vs conventional fibrino-
ment (PRISM) Study Investigators. A comparison of aspirin lytic therapy for acute myocardial infarction: GUSTO V
plus tirofiban with aspirin plus heparin for unstable angina. randomized trial. JAMA 2002; 288:2130 –2135
N Engl J Med 1998; 338:1498 –1505 331 Abciximab Emergent Stroke Treatment Trial (AbESTT)
325 Platelet Receptor Inhibition in Ischemic Syndrome Manage- Investigators. Emergency administration of abciximab for
ment in Patients Limited by Unstable Signs and Symptoms treatment of patients with acute ischemic stroke: results of a
(PRISM-PLUS) Study Investigators. Inhibition of the plate- randomized phase 2 trial. Stroke 2005; 36:880 – 890
let glycoprotein IIb/IIIa receptor with tirofiban in unstable 332 Bertolucci AA, Howard G. Meta-analysis of data from the six
angina and non-Q-wave myocardial infarction. N Engl primary prevention trials of cardiovascular events using
J Med 1998; 338:1488 –1497 aspirin. Am J Cardiol 2006; 98:746 –750

Antiplatelet Drugs* : American College of Chest Physicians
Evidence-Based Clinical Practice Guidelines (8th Edition)
Carlo Patrono, Colin Baigent, Jack Hirsh and Gerald Roth
Chest 2008;133; 199S-233S
DOI 10.1378/chest.08-0672
This information is current as of March 22, 2010
Updated Information Updated Information and services, including

& Services high-resolution figures, can be found at:
http://chestjournal.chestpubs.org/content/133/6_suppl/1
99S.full.html
References This article cites 328 articles, 183 of which can be
accessed free at:
http://chestjournal.chestpubs.org/content/133/6_su
ppl/199S.full.html#ref-list-1
Citations This article has been cited by 14 HighWire-hosted
articles:
http://chestjournal.chestpubs.org/content/133/6_su
ppl/199S.full.html#related-urls
Open Access Freely available online through CHEST open access
option
Permissions & Licensing Information about reproducing this article in parts
(figures, tables) or in its entirety can be found online at:
http://www.chestjournal.org/site/misc/reprints.xhtml
Reprints Information about ordering reprints can be found online:
http://www.chestjournal.org/site/misc/reprints.xhtml
Email alerting service Receive free email alerts when new articles cite this
article. Sign up in the box at the top right corner of the
online article.
Images in PowerPoint Figures that appear in CHEST articles can be
format downloaded for teaching purposes in PowerPoint slide
format. See any online article figure for directions


Anti Platelet Drugs Accp 8th Ed

Uploaded by

Copyright:

Available Formats

Anti Platelet Drugs Accp 8th Ed

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Anti Platelet Drugs Accp 8th Ed

Uploaded by

Copyright:

Available Formats

Antiplatelet Drugs* : American College of

Chest Physicians Evidence-Based Clinical

The online version of this article, along with updated information

CHEST is the official journal of the American College of Chest

Downloaded from chestjournal.chestpubs.org by guest on March 22, 2010

P latelets are vital components of normal hemosta-

www.chestjournal.org CHEST / 133 / 6 / JUNE, 2008 SUPPLEMENT 199S

Downloaded from chestjournal.chestpubs.org by guest on March 22, 2010

200S Antithrombotic and Thrombolytic Therapy 8th Ed: ACCP Guidelines

Downloaded from chestjournal.chestpubs.org by guest on March 22, 2010

www.chestjournal.org CHEST / 133 / 6 / JUNE, 2008 SUPPLEMENT 201S

Downloaded from chestjournal.chestpubs.org by guest on March 22, 2010

Percentage Inhibition of Urinary 2,3-

dinor-TXB2 Excretion In Vivo

202S Antithrombotic and Thrombolytic Therapy 8th Ed: ACCP Guidelines

Downloaded from chestjournal.chestpubs.org by guest on March 22, 2010

www.chestjournal.org CHEST / 133 / 6 / JUNE, 2008 SUPPLEMENT 203S

Downloaded from chestjournal.chestpubs.org by guest on March 22, 2010

Single Duration of Prostanoid Cumulative Effects Upon Daily

Platelets COX-1 100 24–48 Yes 50–81

204S Antithrombotic and Thrombolytic Therapy 8th Ed: ACCP Guidelines

Downloaded from chestjournal.chestpubs.org by guest on March 22, 2010

www.chestjournal.org CHEST / 133 / 6 / JUNE, 2008 SUPPLEMENT 205S

Downloaded from chestjournal.chestpubs.org by guest on March 22, 2010

206S Antithrombotic and Thrombolytic Therapy 8th Ed: ACCP Guidelines

Downloaded from chestjournal.chestpubs.org by guest on March 22, 2010

www.chestjournal.org CHEST / 133 / 6 / JUNE, 2008 SUPPLEMENT 207S

Downloaded from chestjournal.chestpubs.org by guest on March 22, 2010

Table 4 —Primary Prevention Trials (Section 2.4.1)

Mean Eligible Randomized

208S Antithrombotic and Thrombolytic Therapy 8th Ed: ACCP Guidelines

Downloaded from chestjournal.chestpubs.org by guest on March 22, 2010

www.chestjournal.org CHEST / 133 / 6 / JUNE, 2008 SUPPLEMENT 209S

Downloaded from chestjournal.chestpubs.org by guest on March 22, 2010

210S Antithrombotic and Thrombolytic Therapy 8th Ed: ACCP Guidelines

Downloaded from chestjournal.chestpubs.org by guest on March 22, 2010

www.chestjournal.org CHEST / 133 / 6 / JUNE, 2008 SUPPLEMENT 211S

Downloaded from chestjournal.chestpubs.org by guest on March 22, 2010

212S Antithrombotic and Thrombolytic Therapy 8th Ed: ACCP Guidelines

Downloaded from chestjournal.chestpubs.org by guest on March 22, 2010

www.chestjournal.org CHEST / 133 / 6 / JUNE, 2008 SUPPLEMENT 213S

Downloaded from chestjournal.chestpubs.org by guest on March 22, 2010

214S Antithrombotic and Thrombolytic Therapy 8th Ed: ACCP Guidelines

Downloaded from chestjournal.chestpubs.org by guest on March 22, 2010

www.chestjournal.org CHEST / 133 / 6 / JUNE, 2008 SUPPLEMENT 215S

Downloaded from chestjournal.chestpubs.org by guest on March 22, 2010

216S Antithrombotic and Thrombolytic Therapy 8th Ed: ACCP Guidelines

Downloaded from chestjournal.chestpubs.org by guest on March 22, 2010

www.chestjournal.org CHEST / 133 / 6 / JUNE, 2008 SUPPLEMENT 217S

Downloaded from chestjournal.chestpubs.org by guest on March 22, 2010

218S Antithrombotic and Thrombolytic Therapy 8th Ed: ACCP Guidelines

Downloaded from chestjournal.chestpubs.org by guest on March 22, 2010

www.chestjournal.org CHEST / 133 / 6 / JUNE, 2008 SUPPLEMENT 219S

Downloaded from chestjournal.chestpubs.org by guest on March 22, 2010

220S Antithrombotic and Thrombolytic Therapy 8th Ed: ACCP Guidelines

Downloaded from chestjournal.chestpubs.org by guest on March 22, 2010

www.chestjournal.org CHEST / 133 / 6 / JUNE, 2008 SUPPLEMENT 221S

Downloaded from chestjournal.chestpubs.org by guest on March 22, 2010

222S Antithrombotic and Thrombolytic Therapy 8th Ed: ACCP Guidelines

Downloaded from chestjournal.chestpubs.org by guest on March 22, 2010