ARTICLES
Novel dosing regimen of eptifibatide in planned coronary stent
implantation (ESPRIT): a randomised, placebo-controlled trial
The ESPRIT investigators*
Summary
Background The platelet glycoprotein IIb/IIIa inhibitors,
although effective in reducing ischaemic complications of
percutaneous coronary intervention, are used in few
coronary stent implantation procedures. ESPRIT (Enhanced
Suppression of the Platelet IIb/IIIa Receptor with Integrilin
Therapy) is a randomised, placebo-controlled trial to assess
whether a novel, double-bolus dose of eptifibatide could
improve outcomes of patients undergoing coronary stenting.
Methods We recruited 2064 patients undergoing stent
implantation in a native coronary artery. Immediately before
percutaneous coronary intervention, patients were randomly
allocated to receive eptifibatide, given as two 180 g/kg
boluses 10 min apart and a continuous infusion of
2·0 g/kg/min for 18–24 h, or placebo, in addition to
aspirin, heparin, and a thienopyridine. The primary endpoint
was the composite of death, myocardial infarction, urgent
target vessel revascularisation, and thrombotic bailout
glycoprotein IIb/IIIa inhibitor therapy within 48 h after
randomisation. The key secondary endpoint was the
composite of death, myocardial infarction, or urgent target
vessel revascularisation at 30 days.
Findings The trial was terminated early for efficacy. The
primary endpoint was reduced from 10·5% (108 of 1024
patients on placebo [95% CI 8·7–12·4%]) to 6·6% (69 of
1040 [5·1–8·1%]) with treatment (p=0·0015). The key 30
day secondary endpoint was also reduced, from 10·5% (107
of 1024 patients on placebo [8·6–12·3%]) to 6·8% (71 of
1040 [5·3–8·4%]; p=0·0034). There was consistency in
reduction of events across all components of the composite
endpoint and among the major subgroups. Major bleeding
was infrequent but arose more often with eptifibatide than
placebo (1·3%, 13 of 1040 [0·7–2·1%]) vs 0·4%, 4 of 1024
[0·1–1·0%]; p=0·027).
Interpretation Routine glycoprotein IIb/IIIa inhibitor
pretreatment with eptifibatide substantially reduces
ischaemic complications in coronary stent intervention and is
better than a strategy of reserving treatment to the bailout
situation.
Lancet 2000; 356: 2037–44
.see page
*Members listed at end of paper
Correspondence to: Dr James E Tcheng, Box 3275 Duke University
Medical Center, Durham, NC 27710, USA
(e-mail: [email protected])
THE LANCET • Vol 356 • December 16, 2000
Introduction
Platelet glycoprotein IIb/IIIa receptor inhibition offers
clinically significant protection against ischaemic
complications of percutaneous coronary intervention
(PCI),1–6 and improves clinical outcomes for up to 3 years
after PCI.7 Eptifibatide is a parenteral cyclic heptapeptide
with a short half-life and high specificity for glycoprotein
IIb/IIIa integrin.8 In the Integrilin to Minimize Platelet
Aggregation and Coronary Thrombosis-II (IMPACT II)
trial, treatment (analysed on a treated as randomised
basis) with a 135 g/kg bolus of eptifibatide and a
0·5 g/kg/min infusion for 20–24 h reduced the 30 day
composite rate of death, myocardial infarction, and urgent
coronary revascularisation after PCI from 11·6% to 9·1%,
a reduction of 22% (p=0·035).4 These results, however,
were disappointing when compared with those predicted
by antecedent pharmacodynamic dosing studies.9,10
Phillips and colleagues11 subsequently showed that
eptifibatide regimens tested in IMPACT II achieved only
30–50% of maximum blockade of platelet glycoprotein
IIb/IIIa integrin. This percentage was considerably less
than the 80% level of blockade required to prevent
occlusive thrombus formation with the abciximab
precursor m7E3 in dogs,12 and studied in clinical trials of
abciximab in PCI.13 Further dosing studies, coupled with
pharmacodynamic modelling, suggested that a novel,
high-dose, double bolus, 180/2·0/180 (180 g/kg bolus of
eptifibatide followed by a 2·0 g/kg/min infusion for
18–24 h, with a second 180 g/kg bolus 10 min after the
first) regimen would uniformly attain and maintain
greater than 80% blockade of the glycoprotein IIb/IIIa
receptor.14,15 We therefore designed the Enhanced
Suppression of the Platelet IIb/IIIa Receptor with
Integrilin Therapy (ESPRIT) trial to determine the
efficacy and safety of this high-dose regimen of
eptifibatide as an adjunct to planned coronary stent
implantation.16
Methods
Study population
ESPRIT was a randomised, double-blind, placebocontrolled, crossover-permitted, parallel-group clinical
trial done at 92 centres in USA and Canada. Patients with
coronary artery disease scheduled to undergo PCI with
stent implantation in a native coronary artery, and who, in
the opinion of the treating physician, would not routinely
be treated with a glycoprotein IIb/IIIa inhibitor during
PCI, were considered for inclusion. Exclusion criteria
included: myocardial infarction within 24 h before
randomisation; continuing chest pain precipitating urgent
referral for PCI; PCI within the previous 90 days;
previous stent implantation at the target lesion;
anticipated staged PCI in the 30 days after randomisation;
treatment with a glycoprotein IIb/IIIa inhibitor or a
thienopyridine in the 30 days before randomisation (but
with thienopyridine administration required per protocol
on the day of randomisation); stroke or transient
ischaemic attack within 30 days before randomisation; any
history of haemorrhagic stroke; history of bleeding
diathesis or evidence of abnormal bleeding within 30 days
2037
For personal use only. Not to be reproduced without permission of The Lancet.
ARTICLES
before randomisation; major surgery within the previous 6
weeks; uncontrolled hypertension with a systolic blood
pressure greater than 200 mm Hg or diastolic greater than
110 mm Hg; documented thrombocytopenia with a
platelet count less than 100109/L; or a serum creatinine
greater than 350 mol/L. The protocol was approved by
each of the respective institutional review boards, and
informed consent was obtained from all patients.
Study protocol
Patients were pretreated with aspirin and a thienopyridine
(either ticlopidine or clopidogrel, with use of a loading
dose) on the day of randomisation, and study drug was
started immediately before initiation of PCI. The PCI
procedure was done to local standards. Any stent type
with regulatory agency approval could be implanted.
Patients were allocated to receive either placebo or
eptifibatide in a one to one distribution. Randomisation
allocation was done in the catheterisation laboratory after
coronary angiography confirmed the plan to proceed to
PCI. In patients with a totally occluded target lesion, the
occlusion was crossed with a guidewire before
randomisation allocation (the time when any study drug
was first given).
Eptifibatide was delivered as two boluses and an
infusion. The first bolus of 180 g/kg was immediately
followed by initiation of a 2·0 g/kg/min (or 1·0
g/kg/min in patients with serum creatinine values greater
than 177 mol/L) continuous infusion. A second bolus of
180 g/kg was given 10 min after the first. The infusion
was continued until hospital discharge or up to 18–24 h,
whichever took place first. PCI device activation was
allowed any time after the first bolus. A weight adjusted
heparin regimen was recommended (initial bolus of 60
units/kg, not to exceed 6000 units) to target an activated
clotting time between 200 and 300 s. However, heparin
dosing was not controlled by an unblinded heparin
coordinator and was left to the discretion of the treating
physician. Further heparin, after PCI, was discouraged.
Vascular access sheaths were removed, with a femoral
arteriotomy closure device or with external compression
for haemostasis, within 3–4 h of PCI procedure
completion. Blood samples were taken from every patient
within the 4 h preceding randomisation, and every 6 h
thereafter, up to 24 h or hospital discharge (whichever
came first), to assay for markers of myocardial necrosis.
These samples were analysed at Covance (Indianapolis,
IN, USA), the cardiac enzyme core laboratory for the
trial.
Bailout glycoprotein IIb/IIIa inhibitor therapy
To provide emergency open-label glycoprotein IIb/IIIa
inhibitor therapy (for the direct treatment of abrupt
closure, no reflow, coronary thrombosis, or other similar
PCI complications), bailout kits were supplied.
Dependent on patient randomisation, the kit contained
two bolus vials of either eptifibatide (for patients allocated
to placebo) or placebo (for those allocated to eptifibatide).
Once bailout treatment was underway, study drug
infusion was discontinued and switched to an infusion of
open-label eptifibatide. Bailout glycoprotein IIb/IIIa
inhibitor use did not result in unblinding of study drug
assignment.
Study endpoints
The primary clinical efficacy endpoint was the composite
of death, myocardial infarction, urgent target vessel
revascularisation, and thrombotic bailout glycoprotein
IIb/IIIa inhibitor therapy within 48 h after randomisation.
2038
The key secondary efficacy endpoint was the composite of
death, myocardial infarction, and urgent target vessel
revascularisation within 30 days after randomisation. An
independent clinical events committee, blinded to study
group assignment, adjudicated suspected clinical events.
Two types of myocardial infarction were defined as
endpoint events. An enzymatic myocardial infarction
arose when two or more values of the creatine kinasemuscle brain (CK-MB) isoenzyme, for the first 24 h after
PCI, were at least three times the upper limit of normal. A
clinical myocardial infarction was one reported by an
investigator and adjudicated as an endpoint by the clinical
events committee. This latter type of endpoint myocardial
infarction required corroboration in the form of a clinical
syndrome consistent with myocardial infarction, and
supportive electrocardiographic or cardiac marker data.
Supportive electrocardiographic findings included
development of Q waves (at least 0·04 s), in two or more
contiguous leads, or new left-bundle branch block.
Supportive cardiac marker findings included, in order of
precedence and to the exclusion of the next value, raised
CK-MB, troponin I or T, or total creatine kinase
concentrations to at least twice the upper limit of normal.
In specific cases of repeat PCI or coronary artery bypass
surgery, cardiac markers at least three times or five times
the upper limit of normal were required, respectively.
Urgent target vessel revascularisation included any
coronary artery bypass graft surgery, or a second PCI of
the original target vessel, done on a non-elective basis for
recurrent myocardial ischaemia. Similarly, cases who
received bailout glycoprotein IIb/IIIa inhibitor therapy
were adjudicated as an endpoint if treatment was given to
manage a thrombotic complication. Included in this
definition was the development of a new filling defect or
haziness consistent with thrombosis, abrupt closure, or
any other coronary intervention complications felt to be
related to thrombus formation.
Measures of safety included rates of major bleeding,
blood transfusions, and stroke through 48 h or hospital
discharge, whichever arose first. Bleeding events were
quantitatively classified as major or minor according to
the criteria of the Thrombolysis In Myocardial Infarction
(TIMI) study group.17 To account for the effect of red-cell
transfusions on measured haemoglobin values, estimated
decreases in haemoglobin were adjusted according to
Landefeld and colleagues’ technique.18 Bleeding was also
classified qualitatively by the investigator with the Global
Use of Strategies To Open occluded coronary arteries
(GUSTO) classification.19
Data management and statistical analysis
Sample size calculation was based on predicted reduction
in rate of the key 30 day secondary composite efficacy
endpoint, rather than primary endpoint. The key 30 day
endpoint was used in this calculation because of its use in
other trials of platelet glycoprotein IIb/IIIa inhibition. We
predicted an 11% adverse clinical event rate in the
placebo arm, and expected a 33% reduction in this rate.
With 86% power and an alpha of 0·05 (two-tailed), the
projected sample size for the study was 2400 patients.
The randomisation allocation code was generated with
random permuted blocks within each investigative site,
with one to one allocation of treatments. Study drugs were
packaged to be indistinguishable, irrespective of content.
Any drug kit that became unusable was replaced with
another uniquely numbered kit of the same treatment.
This kit was then given to the next patient enrolled at that
site. To ensure patient safety, the trial design allowed
unblinding of the treatment allocation if dictated by a
THE LANCET • Vol 356 • December 16, 2000
For personal use only. Not to be reproduced without permission of The Lancet.
ARTICLES
clinical emergency. An independent data and safety
monitoring board was established to monitor safety of the
trial. Because we anticipated that the recruitment period
would be complete in less than 6 months, interim analyses
of efficacy were not preplanned. However, when the
recruitment period extended beyond 8 months, the data
safety monitoring board independently chose to assess
both safety and efficacy. For this analysis, the board
prespecified a p<0·005 for a reduction in rate of death or
myocardial infarction at 48 h as criteria warranting
termination of the trial.
All analyses were by intention to treat. Efficacy analyses
were pairwise comparisons of the eptifibatide treatment
group and the placebo group by means of conventional 2
analyses. For the secondary analyses, the groups were
compared by 2 analyses for categorical variables, and
Wilcoxon rank-sum tests for continuous variables. The
time to first occurrence for any component of the
composite end-point is displayed with Kaplan-Meier
survival curves. All secondary analyses were prespecified
in the protocol. Continuous variables are presented as
medians with first and third quartiles. Treatment effects
by subgroup are shown as odds ratios with 95% CIs.
Almost all individuals received clopidogrel, a
thienopyridine. Over twenty stent types were used; at least
20% of the total number of stents implanted were
manufactured by each of Medronic/AVE (Santa Rosa,
CA, USA), Guidant/ACS (Santa Clara, CA, USA), and
SciMed/Boston Scientific (Maple Grove, MN, USA).
There were no Palmaz-Schatz stents (Johnson and
Johnson Interventional Systems, Morristown, NJ, USA)
implanted. Study drug was discontinued before 12 h in
9·3% (192) and before 16 h in 11·7% (242), which was
balanced between randomisation groups. There were 35
patients in the eptifibatide group and 43 patients in the
placebo group who crossed over to open-label eptifibatide
therapy. The primary composite clinical endpoint of
death, myocardial infarction, urgent target vessel
revascularisation, or crossover bailout therapy for
thrombotic complications at 48 h, occurred in 108 of the
1024 placebo treated patients and in 69 of the 1040
2064 patients randomised
Results
Enrolment took place between June 3, 1999, and
February 4, 2000. The data and safety monitoring board
recommended that with 2064 patients randomised we
should end enrolment prematurely for efficacy. This
recommendation was based on an interim analysis of
1758 patients with data complete to 48 h, and 1384
patients with completed 30 day data. A 43% relative risk
reduction in irreversible endpoints of death or myocardial
infarction at 48 h (95% CI 19%–61%; p=0·0017) was the
primary determinant of recommendation for termination.
Other prerequisite conditions of the board included
consistency of reduction in death or myocardial infarction
at 30 days, consistency across other components of the
endpoint at 48 h and 30 days, absence of safety concerns,
and results of other clinical trials of glycoprotein IIb/IIIa
inhibition in PCI.
Figure 1 shows the flow of patients through the trial.
Among the 2064 patients finally enrolled in the study,
baseline demographic and angiographic characteristics
were closely similar (table 1). 372 of 2064 (18%) patients
were classified as having an acute coronary syndrome
within 48 h, or an acute ST-segment elevation myocardial
infarction, within 7 days before the intervention. Specific
treatment details are listed in table 2. Most of the patients
enrolled underwent PCI and had at least one stent placed.
1040
eptifibatide
1024
placebo
1040
treated
1024
treated
1025
PCI
15
no PCI
35 bailed
out
986 stent
39 no stent
1 lost
1015
PCI
9
no PCI
43 bailed
out
997 stent
18 no stent
1 lost
1039 with 30-day follow-up
1023 with 30-day follow-up
Figure 1: Trial profile
Characteristic
Eptifibatide (n=1040)
Placebo (n=1024)
Total (n=2064)
Median age (years) (25–75 quartiles)
Median weight (kg) (25–75 quartiles)
Women
USA/Canada
Previous myocardial infarction
Previous PCI
Previous CABG
Diabetes
Hypertension
Hypercholesterolaemia
Smoker
Stable angina
UA/NQMI (2–180 days)
UA/NQMI within 2 days
ST elevation myocardial infarction (within 7 days)
Positive functional test only
Other anginal equivalent
62 (54–71)
84·0 (73·9–95·3)
280 (27%)
772 (74%)/268 (26%)
331 (32%)
237 (23%)
106 (10%)
208 (20%)
608 (59%)
600 (58%)
250 (24%)
407 (39%)
331 (32%)
139 (13%)
44 (4%)
96 (9%)
23 (2%)
62 (54–71)
84·7 (75·0–97·0)
282 (28%)
759 (74%)/265 (26%)
321 (31%)
246 (24%)
105 (10%)
211 (21%)
605 (59%)
599 (59%)
228 (23%)
387 (38%)
333 (33%)
140 (14%)
49 (5%)
91 (9%)
24 (2%)
62 (54–71)
84·4 (74·0–96·0)
562 (27%)
1531 (74%)/533 (26%)
652 (32%)
483 (23%)
211 (10%)
419 (20%)
1213 (59%)
1199 (58%)
478 (23%)
794 (39%)
664 (32%)
279 (14%)
93 (5%)
187 (9%)
47 (2%)
PCI=percutaneous coronary intervention; CABG=coronary artery bypass grafting; UA=unstable angina; NQMI=non-Q-wave myocardial infarction.
Table 1: Baseline characteristics of patients
THE LANCET • Vol 356 • December 16, 2000
2039
For personal use only. Not to be reproduced without permission of The Lancet.
ARTICLES
Characteristic
Eptifibatide (n=1040)
Median infusion duration (h) (25–75 quartiles)
PCI done
Placebo (n=1024)
18·3 (18·0–20·1)
Total (n=2064)
18·4 (18·0–20·2)
18·4 (18·0–20·2)
1025 (99%)
1015 (99%)
2040 (99%)
986 (95%)
676 (65%)
234 (23%)
54 (5%)
22 (2%)
997 (97%)
650 (64%)
242 (24%)
90 (9%)
15 (2%)
1983 (96%)
1326 (64%)
476 (23%)
144 (7%)
37 (2%)
Femoral arteriotomy closure device
189 (18%)
167 (16%)
356 (17%)
Median max ACT (s) (25–75 quartiles)
273 (234–316)
263 (230–302)
Number of stents placed
1
2
3
4
Thienopyridine use
1009 (97%)
268 (232–309)
1006 (98%)
2015 (98%)
PCI=percutaneous coronary intervention; Max ACT=maximum procedural activated clotting time.
Table 2: Treatment indices
eptifibatide treated patients, a risk ratio of 0·63 (95%
CI 0·47–0·84), from 10·5% (8·7–12·4%) to 6·6%
(5·1–8·1%; p=0·0015). Figure 2 shows event rates for
individual components and combinations of components
of the endpoint. All components and combinations of
components of the primary endpoint were reduced in
frequency to about the same degree as the primary
composite endpoint.
Figure 3 shows reductions in the rate of the 48 h
primary composite clinical endpoint among subgroups.
Benefit was realised irrespective of baseline characteristics
including age, weight, sex, diabetes, and disease
presentation. The degree of heparin anticoagulation, as
measured by the activated clotting time, did not predict
events. In the placebo group, the frequency of the primary
composite endpoint was closely similar among patients
when divided into tertiles of maximum activated clotting
time (10·0%, 95% CI 6·9–13·2%, in the lowest tertile;
11·5%, 8·1–14·8%, in the middle tertile; and 10·3%,
6·8–13·7%, in the highest tertile). The benefit of
treatment with eptifibatide was also similar across the
tertiles of activated clotting time, with the lowest overall
event rate of the primary composite endpoint occurring in
patients receiving eptifibatide who were in the lowest
tertile.
The key secondary composite endpoint of death,
myocardial infarction, and urgent target vessel
revascularisation at 30 days occurred in 71/1040 patients
(68%, 95% CI 5·3–8·4%) treated with eptifibatide and
107/1040 of those treated with placebo (10·5%, 8·6–12·3%
[p=0·0034]), a risk ratio of 0·65 (95% CI 0·49–0·87).
Figure 4 shows Kaplan-Meier plots of events to 30 days.
After the first 48 h, curves for death, myocardial infarction,
and urgent target vessel revascularisation, and for death
and myocardial infarction were fairly flat. Over 88%
(157/178) of all events occurred in the first 48 h. There
were no rebound or other clustering of events recorded at
the time of discontinuation of treatment.
Rates of severe and moderate bleeding were much the
same between groups (table 3). However, as measured by
the quantitative TIMI trial bleeding criteria, there was an
increase in bleeding associated with treatment. When
TIMI bleeding was analysed by tertile of the activated
clotting time, rates were not increased in the lowest tertile:
0·6% (two of 316, 95% CI 0·08–2·2%) with placebo
versus 0·7% (two of 287, 0·08–2·5%) with eptifibatide.
Rates of bleeding increased as the activated clotting time
rose, with bleeding being enhanced by treatment with
eptifibatide. There were two cases of acute profound
thrombocytopenia in the eptifibatide group (a decrease in
the platelet count to <20109/L within 24 h of start of
treatment). Both were managed conservatively without
platelet transfusion; thrombocytopenia began to resolve
within hours of discontinuation of eptifibatide, and there
were no clinical sequelae. Finally, rates of red blood cell
transfusion were low (table 3).
Eptifibatide
Placebo
RR
95% CI
p
Primary endpoint
6·6%
10·5%
0·63%
0·47–0·84
0·0015
Death/MI/UTVR
6·0%
9·3%
0·65%
0·47–0·87
0·0045
Death/MI
5·5%
9·2%
0·60%
0·44–0·82
0·0013
Death/large MI
3·4%
5·1%
0·67%
0·44–1·01
0·053
Large MI
3·3%
4·9%
0·67%
0·44–1·03
0·064
All MI
5·4%
9·0%
0·60%
0·44–0·83
0·0015
UTVR
0·6%
1·0%
0·60%
0·22–1·62
0·30
Thrombotic bailout
1·0%
2·1%
0·48%
0·21–0·94
0·029
0·1%
0·2%
0·50%
0·05–5·42
0·55
5·4
Death
0·5
1
1·5
Eptifibatide better
Placebo better
Figure 2: Risk ratios (95% CI) for major endpoint components at 48 h
0
2
RR=risk ratio; UTVR=urgent target vessel revascularisation; MI=myocardial infarction; large MI=CK-MB more than five times upper limit of normal.
2040
THE LANCET • Vol 356 • December 16, 2000
For personal use only. Not to be reproduced without permission of The Lancet.
ARTICLES
Eptifibatide
Age (years)
65
65
Placebo
RR
95% CI
p
6·8%
6·5%
8·1%
13·7%
0·84%
0·47%
0·56–1·25
0·31–0·72
0·38
0·0003
7·9%
5·3%
6·6%
14·1%
9·9%
8·0%
0·56%
0·54%
0·82%
0·36–0·87
0·31–0·93
0·48–1·40
0·009
0·024
0·47
Weight (kg)
Lower tertile
Middle tertile
Upper tertile
Diabetes
Yes
No
SEX
Male
Female
3·9%
7·3%
6·6%
11·6%
0·58%
0·63%
0·25–1.35
0·47–0·86
0·20
0·0033
6·8%
6·1%
9·0%
14·5%
0·76%
0·42%
0.54–1·07
0·24–0·72
0·12
0·001
ACT
Lower tertile
Middle tertile
Upper tertile
6·1%
7·3%
7·0%
10·0%
11·5%
10·3%
0·63%
0·63%
0·68%
0·36–1·04
0·39–1·03
0·41–1·12
0·065
0·06
0·12
CAD presentation
Stable angina
ACS 2 days
ACS 2 days
ST elev MI 7 days
5·4%
5·7%
7·9%
11·4%
7·2%
11·1%
15·0%
20·4%
0·75%
0·52%
0·53%
0·56%
0·44–1·28
0·30–0·88
0·26–1·05
0·21–1·50
0·29
0·013
0·063
0·24
0
0·5
Eptifibatide better
1
1·5
Placebo better
2
Figure 3: Subgroup analysis of the primary endpoint including risk ratios (95% CI)
ACT=activated clotting time; CAD=coronary artery disease; ACS=acute coronary syndrome; ST elevation=ST segment elevation; RR=risk ratio. The weight tertiles
for men were <81 kg, 81–95 kg, and >95 kg, and for women <68 kg, 68–82 kg, and >82 kg.
Characteristic
100%
12%
Placebo
10·5%
10%
8%
Eptifibatide
6·8%
6%
Cumulative event rate (%)
4%
2%
Placebo
(n=1024)
Total
(n=2064)
Intracranial bleeding
2 (0·2%)
1 (0·1%)
3 (0·1%)
Non-haemorrhagic stroke
1 (0·1%)
0
Severe bleeding*
7 (0·7%)
5 (0·5%)
Moderate bleeding*
14 (1%)‡
Major bleeding (TIMI)†
Overall
ACT <244 (n=661)
ACT 244–292 (n=678)
ACT >292 (n=676)
13 (1%)‡
2/312 (0·6%)
6/330 (1·8%)
5/374 (1·3%)
Minor bleeding (TIMI)†
29 (2·8%)
Platelet count <20 000
2 (0·2%)
RBC transfusion§
RR= 0·65%
p = 0·0030
0%
Eptifibatide
(n=1040)
15 (1%)
11 (1%)
1 (<0·1%)
12 (0·6%)
25 (1%)
4 (0·4%)‡
17 (1%)
2/349 (0·6%) 4/661 (0·6%)
1/348 (0·3%) 7/678 (1·0%)
1/302 (0·3%) 6/676 (0·9%)
18 (1·7%)
0
10 (1%)
47 (2·3%)
2 (0·1%)
25 (1%)
RBC=red blood cell; ACT=activated clotting time. *Bleeding rates as determined by site
investigator. †See text for explanation of TIMI bleeding classification. ‡p=0·027.
§Includes CABG-related transfusions of red blood cells.
Table 3: Stroke and bleeding complications
100%
12%
Placebo
10%
10·2%
8%
Eptifibatide
6·4%
6%
The pharmacodynamic properties of the eptifibatide
dose tested in ESPRIT achieved: blockade of greater than
90% of available glycoprotein IIb/IIIa receptors in more
than 90% of patients; greater than 90% inhibition of ex
vivo platelet aggregation after stimulation by 20 mol/L
adenosine diphosphate with PPACK (D-Phe-ProArg-CH2CL) as an anticoagulant; and avoidance of
rebound in the first 30 min to 1 h, as occurs after a single
bolus.14
4%
Discussion
2%
RR= 0·62%
p = 0·0014
0%
0
10
20
Days from randomisation
30
Figure 4: Kaplan-Meier plot of probability of composite endpoint
of death, myocardial infarction (upper), and urgent TVR and
death and myocardial infarction (lower) to 30 days.
TVR=target vessel revascularisation; RR=risk ratio.
THE LANCET • Vol 356 • December 16, 2000
There are several key findings of the ESPRIT trial. First,
treatment with a novel, high-dose, double-bolus regimen
of eptifibatide, as an adjunct to contemporary stent
implantation, produced significant and clinically relevant
reductions in ischaemic complications of this procedure.
These benefits were achieved despite exclusion of high
risk patients; patients enrolled in ESPRIT did not warrant
glycoprotein IIb/IIIa receptor blockade as a pretreatment.
Additionally, results were achieved under conditions of
typical contemporary PCI—namely, the implantation of
2041
For personal use only. Not to be reproduced without permission of The Lancet.
ARTICLES
stents deployed at high balloon pressures along with
modern adjunctive drug treatment, particularly the
universal use of thienopyridines and low-dose heparin.
Furthermore, outcomes were realised by adding
eptifibatide to standard algorithms of PCI patient care.
Our results were consistent across individual components
of the composite primary clinical endpoint as well as
across combinations of components of the primary
endpoint. This consistency of benefit was also seen across
subgroups of patients independent of baseline
characteristics. Treatment effects were maintained to 30
days. Finally, although there were trends toward increased
rates of bleeding, our findings suggest that safety of
eptifibatide could be enhanced by careful control of
heparin dosing during PCI.
The ESPRIT trial used a higher dose of a glycoprotein
IIb/IIIa antagonist than previous studies, and thus
explored whether greater inhibition of platelet aggregation
enhanced the magnitude of benefit. Studies conducted
after IMPACT II showed that ex vivo measurements of
the pharmacodynamics of eptifibatide were dependent on
choice of blood anticoagulant; with the doses of
eptifibatide tested in IMPACT II, only moderate (about
50%) values of glycoprotein IIb/IIIa receptor blockade,
resulting in only 50–60% inhibition of platelet
aggregation, were achieved.9 Even with the differences in
study design, year of conduct, and PCI technique, the
ESPRIT results support the hypothesis that the degree of
inhibition of platelet glycoprotein IIb/IIIa integrin
correlates with the potential for clinical benefit of PCI.
Our findings of greater efficacy, with relative respect to
that in the IMPACT II trial, lend support to the following
notions. First, the clinical benefit of glycoprotein IIb/IIIa
inhibition during PCI is directly related to the degree of
inhibition of platelet aggregation and, second, as a
corollary, maximum inhibition of platelet aggregation is
needed to achieve the highest clinical efficacy.
The past 2–3 years have witnessed an explosion in the
development and commercial availability of new stent
design. Also, major shifts in deployment technique
(notably the use of high-pressure inflation strategies) and
adjunctive treatment with the thienopyridines have greatly
reduced procedural complication rates of PCI. In view of
these advantages, stent implantation has become the
technique of choice in PCI and is used in over 80% of all
procedures.20 The advantages of stent implantation,
however, seem to be largely restricted to reducing the
need for acute and delayed repeat intervention; published
work has not shown a benefit over balloon angioplasty in
reducing subsequent death or myocardial infarction.21–23
Our results show that even with the most recent advances
in stent technology, overall rates of death or myocardial
infarction, without glycoprotein IIb/IIIa treatment,
remain the same as in previous reports. Furthermore,
glycoprotein IIb/IIIa inhibition with eptifibatide provides
reductions in major ischaemic complications of
percutaneous revascularisation, even in the face of these
technological advances. ESPRIT is thus a contemporary
study designed to assess incremental benefit of
glycoprotein IIb/IIIa inhibition over modern treatment
approaches that feature universal stent deployment,
smaller and longer stents, low-dose heparin, and
adjunctive treatment with thienopyridines.
In the Evaluation of IIb/IIIa Platelet Inhibition for
Stenting (EPISTENT) trial, implantation of the PalmazSchatz stent with heparin alone (to achieve an activated
clotting time of over 300 s) was compared with stent
implantation with abciximab and low-dose heparin
(activated clotting time target of 200–300 s) and to
2042
balloon angioplasty (with bailout stenting allowed) with
abciximab and low-dose heparin (activated clotting time
target of 200–300 s).6 The findings of ESPRIT seem in
accord with those of EPISTENT. In view of trial
differences, particularly with regard to patient
characteristics (and exclusion from ESPRIT of patients in
whom glycoprotein IIb/IIIa pretreatment was indicated),
study design (with crossover treatment allowed in
ESPRIT), and adjunctive therapies (thienopyridine use
and a lower target activated clotting time range), absolute
differences in clinical events and risk ratios were more
similar than different between the trials. In fact, among
patients in both trials who received aspirin, heparin, and a
thienopyridine, the addition of glycoprotein IIb/IIIa
blockade achieved closely similar reductions in the
composite of death, myocardial infarction, or urgent
target vessel revascularisation at 30 days (10·5–6·9% [risk
ratio 0·67, with an absolute difference of 3·6%] in
ESPRIT and 8·9–5·2% [0·58, with an absolute difference
of 3·7%] in EPISTENT).24 Moreover, inclusion of a
bailout strategy is likely to have reduced placebo event
rate in ESPRIT by ending threatened thrombotic
complications. As in EPISTENT: the frequency of the 30
day composite event rate in ESPRIT was between 10%
and 11%; the rate of the composite endpoint and
components thereof were reduced with treatment; most of
the treatment benefit was realised within the first 48 h; the
benefit of the short course of treatment in hospital was
maintained to 30 days; individual subgroups of patients
responded to treatment consistent with the overall study;
and relative benefit tended to increase as risk increased
(particularly in patients with recent acute coronary
syndromes).
Despite the salutary benefit shown in EPISTENT and
four other clinical trials of glycoprotein IIb/IIIa integrin
blockade with abciximab in PCI,1–3 rate of use of
abciximab in USA has remained at less than 25% of
patients undergoing PCI,25 and in the rest of the world is
even lower. The most common reason cited for avoiding
use of abciximab is drug acquisition cost (about
US$1400–1500 per treatment).26 Although a formal
economic analysis is still to be completed, the difference
in baseline cost of eptifibatide (about US$400–500 per
treatment) and abciximab is similar to the net cost of
abciximab treatment 1 year after stent placement.27 Our
findings, particularly if they are maintained over 6–12
months, should have substantial economic and
therapeutic implications; with a reduced economic barrier
to entry, treatment benefit can be provided to a greater
number of patients.
Heparin dosing was not controlled between groups in
the ESPRIT trial. From the standpoint of efficacy, degree
of heparin anticoagulation did not correlate with
outcome. When divided into tertiles of activated clotting
time, event rates with placebo treatment were similar
across the patient groups. Also, relative and absolute
efficacies with eptifibatide treatment were qualitatively
and quantitatively similar across the groups. However,
bleeding correlated directly with the activated clotting
time tertile; the highest rates of bleeding occurring in
the highest tertile. This correlation suggests that the
appropriate peak target activated clotting time should be
between 200 and 250 s when eptifibatide is used in PCI.
Other safety concerns raised in this trial seem minor.
Acute profound thrombocytopenia (to a platelet count of
less than 20109/L) was recorded in 0·2% of patients;
although this rate is substantially lower than the 0·7%
reported with abciximab,28 continued vigilance for
development of thrombocytopenia is warranted.
THE LANCET • Vol 356 • December 16, 2000
For personal use only. Not to be reproduced without permission of The Lancet.
ARTICLES
Our findings argue that glycoprotein IIb/IIIa inhibition
should be considered a standard of preventive care in nonemergency coronary stent intervention. Benefit was
achieved even with near universal thienopyridine therapy
and availability of bailout, open-label treatment in the
placebo group. There thus seems much to be gained with
glycoprotein IIb/IIIa inhibition even in a low to moderate
risk population. In an era of reduced health care
reimbursements, the cost advantages are obvious.
Determination of long-term efficacy and cost effectiveness
will complete the remaining pieces of the picture that
describe the therapeutic utility of this novel agent in stent
PCI.
Writing Committee—J E Tcheng, J C O’Shea, E A Cohen, C M Pacchiana,
M M Kitt, T J Lorenz, S Greenberg, J Strony, R M Califf.
Steering Committee—C Buller, R M Califf, W J Cantor, E A Cohen,
D M Joseph, M M Kitt, A M Lincoff, T J Lorenz, M Madan, J C O’Shea,
C M Pacchiana, J Popma, J Strony, J E Tcheng, P Teirstein.
Study Centres, principal investigators, and study coordinators—
Sunnybrook & Women’s College Health Science Centre, Toronto, ON,
Canada (E Cohen, L Balleza, P Parsons); Jackson-Madison County
General Hospital, Jackson, TN (H Lui, J Young); Vancouver General
Hospital, Vancouver, BC, Canada (C Buller, R Fox); University of
Ottawa Heart Institute, Ottawa, ON, Canada (M Labinaz, J Jelley,
J Williams); Beth-Israel Deaconess Medical Center, Boston, MA
(D Cohen, M Trovato); Saint Vincent Hospital, Erie, PA (J Smith,
P Henry); St. Michael’s Hospital, Toronto, ON, Canada (R Chisholm,
D O’Donnell); University Hospital of Arkansas/Central Arkansas Veterans
Healthcare, Little Rock, AR (J D Talley, R Pacheco); William Beaumont
Hospital, Royal Oak, MI (S Timmis, A Muraka); Wake Heart Associates,
Raleigh, NC (T Mann, G Cubeddu); Mercy Hospital Medical Center,
Des Moines, IA (M Tannenbaum, J Greene); Orlando Heart Center,
Orlando, FL (E Santoian, M Wash); EMH Regional Medical Center,
Elyria, OH (S Sheldon, L Pronesti); South Texas Hospital, San Antonio,
TX (A Jain, M Alonzo); The Toronto Hospital, Toronto, ON, Canada
(P Seidelin, J Richards); Charlotte Regional Medical Center,
Port Charlotte, FL (M Lopez, R Dittenber, K Johnson); Baylor,
Ben Taub & VA, Houston, TX (G Levine, K Maresh, T Ferrando);
UVA Health System, Charlottesville, VA (I Sarembock, L Snyder);
JFK Medical Center, Atlantis, FL (J Kieval, L Herlan); Pitt County
Memorial Hospital, Greenville, NC (M Miller, D Bembridge); University
Hospitals of Cleveland, Cleveland, OH (R Nair, L Hickel); Hartford
Hospital, Hartford, CT (F Kiernan, D Murphy, J Cloutier); Memorial
Hospital, Chattanooga, TN (E Conn, J Beardsley); Richard Roudebush
VA Medical Center, Indianapolis, IN (M Ritchie, D Cragen);
University of Pittsburgh Medical Center, Pittsburgh, PA (M Z Jafar,
P Counihan, D Rosenfelder); Health Science Centre, Winnipeg, MB,
Canada (J Ducas, L Montebruno); St Paul’s Hospital, Vancouver, BC,
Canada (R Carere, B Radons, L Williams); St Alphonsus Regional
Medical Center, Boise, ID (W Owens, R Dougal); Monroe Regional
Medical Center, Ocala, FL (R Feldman, D Audrain); St Vincent Medical
Center, Toledo, OH (A Karamali, M Smith); Geisinger Medical Center,
Danville, PA (J Blankenship, S L Demko), Rochester General Hospital,
Rochester, NY (M Thompson, G Gacoich, V Chiodo, P Noll); Albert
Einstein Medical Center, Philadelphia, PA (G Ledley, C Miller);
Covenant Healthcare, Saginaw, MI (W Felten, B Garner); University
Hospital, Augusta, GA (A B Chandler, P Easler); St Mary’s Medical
Center, Duluth, MN (G Albin, A Page); Mission St Joseph’s, Asheville,
NC (W Maddox, S Allen); Penn State University Hospital, Hershey,
PA (I Gilchrist, R Moore, H Zimmerman); Foothills Hospital, Calgary,
AB, Canada (M Curtis, K Hildebrand); Alta Bates, Berkely, CA
(R Greene, E Healy); Northern Michigan Hospital, Petoskey, MI
(W Meengs, D Carson); Western Pennsylvania Hospital, Pittsburgh,
PA (J George, T Roncevich); Kaiser Los Angeles Medical Center,
Los Angeles, CA (V Aharonian, R Browning); London Health Sciences
Ctr-University, London, ON, Canada (W Kostuk, S Carr); New York
University Medical Center, New York, NY (F Feit, B Gostomsky);
University Medical Center, Tucson, AZ (S Butman, E Hannah); Baptist
Medical Center, Jacksonville Beach, FL (C D Hassel, D Hartley); Good
Samaritan Hospital, Los Angeles, CA (T Shook, S Hiller-Mullin);
Washington Adventist Hospital, Tacoma Park, MD (D Brill, M Dillion);
Holston Valley Hospital and Medical Center, Kingsport, TN
(B Armstrong, D Kerns); Washington Hospital Center, Washington, DC
(A Pichard, P Okubagzi); The Care Group, Indianapolis, IN (T Nasser,
L Driver); Western Pennsylvania Hospital, Pittsburgh, PA (J Garrett,
L Boltey); University of Texas Medical School, Galveston, TX
(G Stouffer, M Potter); Overlake Hospital Medical Center, Seattle, WA
(T Amidon, S Eggert); Florida Hospital Medical Center, Orlando, FL
(A Taussig, K Potter); Hamilton Health Sciences Corporation, Hamilton,
ON, Canada (M Natarajan, C Tartaglia); Providence Seattle, Seattle, WA
(J Werner, S Eggert); Santa Rosa Memorial Hospital, Santa Rosa, CA
THE LANCET • Vol 356 • December 16, 2000
(T Dunlap, P Harrold-Runge); Northwestern Memorial Hospital,
Chicago, IL (C Davidson, L Goodreau); University of Maryland Medical
Center, Baltimore, MD (W Herzog, N Calamunci); St Luke’s/Roosevelt
Hospital, New York, NY (J Slater, D Tormey); Wilson Regional Medical
Center, Johnson City, NY (W Phillips, D White); London Health
Sciences Ctr-Victoria, London, ON, Canada (K Sridhar, J White);
Scripps Memorial Hospital LaJolla, LaJolla, CA (D Goodman,
M Buchbinder, V Nasser); Alvarado Hospital Medical Center, San Diego,
CA (K Rapeport, P Vorman); University of Massachusetts Medical
Center, Worcester, MA (B Weiner, M Borbone); Presbyterian Heart
Group, Albuquerque, NM (N Shadoff, C Paap); St Francis Hospital,
Roslyn, NY (A Rehman, E Haag); University of Colorado Health Center,
Denver, CO (J Burchenal, K Kioussopoulos); Jewish Hospital, Louisville,
KY (D Senior, J Senior); Brigham and Women’s Hospital, Boston, MA
(R Piana, James Kirshenbaum, S Chan, A Chowdry); Henry Ford
Hospital, Detroit, MI (P Kraft, V Clark, M Fox); New York HospitalCornell University Medical Center, New York, NY (E Deutsch,
T Shannon); Baptist Hospital of Miami, Miami, FL (R Quesada,
J Brotherton, C Murrin); Peninsula Regional Medical Center, Salisbury,
MD (J Cinderella, S Hearne, V Seefried); Allegheny General Hospital,
Pittsburgh, PA (T Farah, D Pakstis); North Mississippi Medical Center,
Tupelo, MS (B Bartolet, C Bond); Deborah Heart and Lung Center,
Browns Mills, NJ (C Debarardinis, D Montory); Methodist Health
Systems, Memphis, TN (G Smith, D Gray); Scripps Mercy Hospital, San
Diego, CA (P Phillips, S Hathaway); Emory University Hospital, Atlanta,
GA (S Manoukian, C Patrick); Grant Riverside Methodist, Columbus,
OH (S Yakubov, J Brooks); Providence St Vincent Medical Center,
Portland, OR (P Block, B Block); Latter Day Saints Hospital, Salt Lake
City, UT (J B Muhlestein, S Kim); Sinai Samaritan Medical Center,
Milwaukee, WI (Y Shalev, W Schmidt); Johns Hopkins Hospital,
Baltimore, MD (J Resar, K Citro); Sentara Norfolk General Hospital,
Norfolk, VA (R Stine, J Zumbuhl); Robert Wood Johnson University
Hospital, New Brunswick, NJ (A Moreyra, S Kreiger).
Data and Safety Monitoring Committee—P Berger, C P Cannon, L Fisher,
V Hasselblad, C M Pacchiana.
Coordinating Centre—W J Cantor, A Foster, V Hasselblad, D Joseph,
M Madan, C McLendon, J C O’Shea, C M Pacchiana, M Rund,
J E Tcheng, N Tillery, F Wood.
Clinical Events Committee—W J Cantor, K W Mahaffey, J C O’Shea.
Covance Core Laboratory—C Irwin, D Kulick, N Johnson.
COR Therapeutics—J Chen, S Greenberg, C Hogeboom, M M Kitt,
T J Lorenz, R Terifay.
Schering-Plough—J Strony, E Veltri.
Acknowledgments
This study was supported by a research grant from COR Therapeutics,
South San Francisco, CA, USA, and Schering-Plough Corporation,
Kenilworth, NJ, USA.
References
1
2
3
4
5
6
7
8
9
Anon. Use of a monoclonal antibody directed against the platelet
glycoprotein IIb/IIIa receptor in high-risk coronary angioplasty: the
EPIC Investigation. N Engl J Med 1994; 330: 956–61.
The EPILOG Investigators. Platelet glycoprotein IIb/IIIa receptor
blockade and low-dose heparin during percutaneous coronary
revascularization. N Engl J Med 1997; 336: 1689–96.
Anon. Randomised placebo-controlled trial of abciximab before and
during coronary intervention in refractory unstable angina: the
CAPTURE Study. Lancet 1997; 349: 1429–35.
Anon. Randomised placebo-controlled trial of effect of eptifibatide on
complications of percutaneous coronary intervention: IMPACT-II.
Integrilin to Minimise Platelet Aggregation and Coronary
Thrombosis-II. Lancet 1997; 349: 1422–28.
The RESTORE Investigators. Effects of platelet glycoprotein IIb/IIIa
blockade with tirofiban on adverse cardiac events in patients with
unstable angina or acute myocardial infarction undergoing coronary
angioplasty: randomised efficacy study of Tirofiban for outcomes and
restenosis. Circulation 1997; 96: 1445–53.
The EPISTENT Investigators. Randomised placebo-controlled and
balloon-angioplasty-controlled trial to assess safety of coronary
stenting with use of platelet glycoprotein-IIb/IIIa blockade: evaluation
of Platelet IIb/IIIa Inhibitor for Stenting. Lancet 1998; 352: 87–92.
Topol EJ, Ferguson JJ, Weisman HF, et al. Long-term protection from
myocardial ischemic events in a randomized trial of brief integrin
beta3 blockade with percutaneous coronary intervention: evaluation of
Platelet IIb/IIIa Inhibition for Prevention of Ischemic Complication.
JAMA 1997; 278: 479–84.
Scarborough RM, Naughton MA, Teng W, et al. Design of potent
and specific integrin antagonists with high specificity for glycoprotein
IIb/IIIa. J Biol Chem 1993; 268: 1066–73.
Tcheng JE, Harrington RA, Kottke-Marchant K, et al. Multicenter,
randomized, double-blind, placebo-controlled trial of the platelet
2043
For personal use only. Not to be reproduced without permission of The Lancet.
ARTICLES
10
11
12
13
14
15
16
17
18
integrin glycoprotein IIb/IIIa blocker Integrelin in elective coronary
intervention. Circulation 1995; 91: 2151–57.
Harrington RA, Kleiman NS, Kottke-Marchant K, et al. Immediate
and reversible platelet inhibition after intravenous administration of a
peptide glycoprotein IIb/IIIa inhibitor during percutaneous coronary
intervention. Am J Cardiol 1995; 76: 1222–27.
Phillips DR, Teng W, Arfsten A, et al. Effect of Ca2+ on GP IIb-IIIa
interactions with integrilin: enhanced GP IIb-IIIa binding and
inhibition of platelet aggregation by reductions in the concentration of
ionized calcium in plasma anticoagulated with citrate. Circulation
1997; 96: 1488–94.
Gold HK, Gimple LW, Yasuda T, et al. Pharmacodynamic study of
F(ab`)2 fragments of murine monoclonal antibody 7E3 directed
against human platelet glycoprotein IIb/IIIa in patients with unstable
angina pectoris. J Clin Invest 1990; 86: 651–59.
Tcheng JE, Ellis SG, George BS, et al. Pharmacodynamics of chimeric
glycoprotein IIb/IIIa integrin antiplatelet antibody Fab 7E3 in high
risk angioplasty. Circulation 1994; 90: 1757–64.
Tcheng JE, Thel MC, Jennings L, et al. Platelet glycoprotein IIb/IIIa
receptor blockade with high-dose Integrilin™ in coronary intervention:
results of the PRIDE study. Eur Heart J 1997; 18: P3615.
Kosoglou T, Belanger B, Jennings L. Eptifibatide pharmacokineticpharmacodynamic relationship in coronary angioplasty patients:
results of the PRIDE study. Eur Heart J 1998; 19 (suppl): 52.
O’Shea JC, Madan M, Cantor WJ, et al. Design and methodology of
the ESPRIT trial: evaluating a novel dosing regimen of eptifibatide in
percutaneous coronary intervention. Am Heart J 2000; 140: 834–39.
Rao AK, Pratt C, Berke A, et al. Thrombolysis in myocardial
infarction (TIMI) trial—phase I: hemorrhagic manifestations and
changes in plasma fibrinogen and the fibrinolytic system in patients
treated with recombinant tissue plasminogen activator and
streptokinase. J Am Coll Cardiol 1988; 11: 1–11.
Landefeld CS, Cook EF, Flatley M, Weisberg M, Goldman L.
2044
19
20
21
22
23
24
25
26
27
28
Identification and preliminary validation of predictors of major
bleeding in hospitalized patients starting anticoagulant therapy. Am J
Med 1987; 82: 703–13.
The GUSTO investigators. An international randomized trial
comparing four thrombolytic strategies for acute myocardial
infarction. N Engl J Med 1993; 329: 673–82.
Belli G, Ellis SG, Topol EJ. Stenting for ischemic heart disease. Prog
Cardiovasc Dis 1997; 40: 159–82.
Rankin JM, Spinelli JJ, Carere RG, et al. Improved clinical outcome
after widespread use of coronary-artery stenting in Canada. N Engl J
Med 1999; 341 (26): 1957–65.
Califf RM, Abdelmeguid AE, Kuntz RE, et al. Myonecrosis
afte revascularization procedures. J Am Coll Cardiol 1998;
31: 241–51.
Abdelmeguid AE, Topol EJ. The myth of the myocardial ‘infarctlet’
during percutaneous coronary revascularization procedures.
Circulation 1996; 94: 3369–75.
Steinhubl SR, Balog C, Topol EJ. Ticlopidine pretreatment prior to
stenting is associated with a substantial decrease in complications:
data from the EPISTENT trial. Circulation 1998; 98 (suppl): I-573.
Jacobs AK, Kip KE, Williams DO, Faxon DP, Cohen HA,
Detre K. Current use of IIb/IIIa receptor antagonists during coronary
intervention: the NHLBI 1997–1998 dynamic registry. Circulation
1998; 98: I-16.
Hillegass WB, Newman AR, Raco DL. Economic issues in
glycoprotein IIb/IIIa receptor therapy. Am Heart J 1999; 138: S24–32.
Topol EJ, Mark DB, Lincoff AM, et al. Outcomes at 1 year and
economic implications of platelet glycoprotein IIb/IIIa blockade in
patients undergoing coronary stenting: results from a multicentre
randomised trial. Lancet 1999; 354: 2019–24.
Berkowitz SD, Harrington RA, Rund MM, Tcheng JE.
Acute profound thrombocytopenia after c7E3 Fab (Abciximab)
therapy. Circulation 1997; 95: 809–13.
THE LANCET • Vol 356 • December 16, 2000
For personal use only. Not to be reproduced without permission of The Lancet.