Akkerhuis KM, Deckers JW, Lincoff AM, Tcheng JE, Boersma E, Anderson K, Balog C, Califf RM, Topol EJ, Simoons ML. Risk of Stroke Associated With Abciximab Among Patients Undergoing Percutaneous Coronary Intervention. JAMA. 2001;286(1):78-82. doi:10.1001/jama.286.1.78
Author Affiliations: Thoraxcenter, University Hospital Rotterdam, Rotterdam, the Netherlands (Drs Akkerhuis, Deckers, Boersma, and Simoons); Cleveland Clinic Foundation, Cleveland, Ohio (Drs Lincoff and Topol and Mr Balog); Duke Clinical Research Institute, Durham, NC (Drs Tcheng and Califf); and Centocor Inc, Malvern, Pa (Dr Anderson).
Context Abciximab, a potent inhibitor of the platelet glycoprotein IIb/IIIa
receptor, reduces thrombotic complications in patients undergoing percutaneous
coronary intervention (PCI). Because of its potent inhibition of platelet
aggregation, the effect of abciximab on risk of stroke is a concern.
Objective To determine whether abciximab use among patients undergoing PCI is
associated with an increased risk of stroke.
Design Combined analysis of data from 4 double-blind, placebo-controlled, randomized
trials (EPIC, CAPTURE, EPILOG, and EPISTENT) conducted between November 1991
and October 1997 at a total of 257 academic and community hospitals in the
United States and Europe.
Patients A total of 8555 patients undergoing PCI with or without stent deployment
for a variety of indications were randomly assigned to receive a bolus and
infusion of abciximab (n = 5476) or matching placebo (n = 3079). One treatment
group in EPIC received a bolus of abciximab only.
Main Outcome Measure Risk of hemorrhagic and nonhemorrhagic stroke within 30 days of treatment
among abciximab and placebo groups.
Results No significant difference in stroke rate was observed between patients
assigned abciximab (n = 22 [0.40%]) and those assigned placebo (n = 9 [0.29%]; P = .46). Excluding the EPIC abciximab bolus-only group,
there were 9 strokes (0.30%) among 3023 patients who received placebo and
15 (0.32%) in 4680 patients treated with abciximab bolus plus infusion, a
difference of 0.02% (95% confidence interval [CI], −0.23% to 0.28%).
The rate of nonhemorrhagic stroke was 0.17% in patients treated with abciximab
and 0.20% in patients treated with placebo (difference, −0.03%; 95%
CI, −0.23% to 0.17%), and the rates of hemorrhagic stroke were 0.15%
and 0.10%, respectively (difference, 0.05%; 95% CI, −0.11% to 0.21%).
Among patients treated with abciximab, the rate of hemorrhagic stroke in patients
receiving standard-dose heparin in EPIC, CAPTURE, and EPILOG was higher than
in those receiving low-dose heparin in the EPILOG and EPISTENT trials (0.27%
vs 0.04%; P = .057).
Conclusions Abciximab in addition to aspirin and heparin does not increase the risk
of stroke in patients undergoing PCI. Patients undergoing PCI and treated
with abciximab should receive low-dose, weight-adjusted heparin.
Intravenous platelet glycoprotein (GP) IIb/IIIa receptor inhibitors
effectively reduce thrombotic complications in patients undergoing percutaneous
coronary intervention (PCI).1- 6
In randomized studies, treatment with abciximab (a monoclonal antibody Fab
fragment directed against the platelet GP IIb/IIIa receptor7)
resulted in an approximate 50% reduction in the composite of death or myocardial
infarction at 30 days.1- 4
However, inhibition of platelet aggregation by GP IIb/IIIa receptor blockers
such as abciximab increases the risk of bleeding complications, particularly
when these agents are combined with conventional doses of heparin.1- 8
Intracerebral hemorrhage is the most serious potential complication
of antithrombotic or anticoagulant therapy and usually results in fatality
or disability.9 Because of its potent inhibition
of platelet aggregation, the effect of abciximab on the risk of stroke has
been a concern. We performed a combined analysis of 8555 patients from 4 large
randomized trials1- 4
to compare stroke rates between patients treated with abciximab (in addition
to aspirin and heparin) and those receiving placebo during PCI.
Data were obtained from EPIC (Evaluation of 7E3 for the Prevention of
Ischemic Complications), CAPTURE (c7E3 Fab Antiplatelet Therapy in Unstable
Refractory Angina), EPILOG (Evaluation in PTCA to Improve Long-term Outcome
with Abciximab GP IIb/IIIa Blockade), and EPISTENT (Evaluation of Platelet
IIb/IIIa Inhibitor for Stenting), 4 studies conducted between November 1991
and October 1997 at a total of 257 academic and community hospitals in the
United States and Europe.
The protocols and results of the 4 studies have been published in detail.1- 4 In brief,
the studies were large, double-blind, placebo-controlled, randomized trials
designed to evaluate the efficacy of the GP IIb/IIIa receptor inhibitor abciximab
in reducing thrombotic complications in patients undergoing PCI for a variety
of indications. The EPIC trial enrolled patients scheduled to undergo balloon
angioplasty in high-risk clinical situations including unstable angina, evolving
myocardial infarction, or high-risk coronary morphology.1
CAPTURE evaluated the use of abciximab in patients with unstable angina refractory
to conventional medical therapy for whom PCI was planned and performed after
approximately 24 hours of pretreatment with abciximab or placebo.2 EPILOG evaluated use of abciximab among patients undergoing
elective balloon angioplasty.3 EPISTENT evaluated
the hypothesis that stenting plus abciximib and balloon angioplasty plus abciximib
would be superior to stenting plus placebo among patients scheduled to undergo
elective or urgent percutaneous coronary revascularization.4
In the EPIC trial (median age [25th and 75th percentiles], 61 [52,68]
years), patients were excluded if they were 80 years of age or older. In EPILOG
(median age, 60 [51,69] years), patients were eligible for inclusion if they
were older than 21 years. In the CAPTURE and EPISTENT trials (median age,
61 and 59 years, respectively), there were no limitations with respect to
age. All trials excluded patients with characteristics associated with an
increased risk of bleeding, as well as those with a cerebrovascular accident
within the preceding 2 years.
Patients were randomly assigned to receive a bolus and an infusion of
abciximab (0.25 mg/kg bolus, 10 µg/min infusion in EPIC and CAPTURE
and 0.125 µg/kg per minute [maximum 10 µg/min] in EPILOG and EPISTENT)
or matching placebo. One treatment group in EPIC received a bolus of abciximab
only. In EPIC, EPILOG, and EPISTENT, the study drug was administered from
1 hour before until 12 hours after PCI. The CAPTURE trial required administration
of placebo or abciximab starting 18 to 24 hours prior to PCI and continuing
until 1 hour after completion of the intervention.
All patients received aspirin and intravenous heparin. With the exception
of EPIC, all trials adjusted the heparin dose for weight for all patients.
In EPIC, heparin was given in an initial bolus of 10 000 to 12 000
U followed by incremental bolus doses to keep the activated clotting time
between 300 and 350 seconds during the intervention.1
CAPTURE and 1 abciximab treatment arm in EPILOG used a standard-dose, weight-adjusted
heparin regimen that consisted of an initial bolus of 100 U of heparin per
kilogram (maximum 10 000 U), with additional boluses as necessary to
achieve and maintain an activated clotting time of at least 300 seconds.2,3 In addition to weight adjustment, patients
in the second abciximab treatment arm in EPILOG and patients in both abciximab
treatment arms in EPISTENT received a low-dose heparin regimen (70 U/kg bolus
with maximum of 7000 U, followed by additional boluses to keep the activated
clotting time at least 200 seconds).3,4
The study protocols were approved by the institutional review board
at each study center and all patients gave written informed consent to participate.
Patients with suspected strokes were identified from the case report
forms for each study. Clinical notes, hospital discharge summaries, neurological
consultation reports, results of computed tomographic or magnetic resonance
imaging studies or reports, and, if applicable, autopsy reports were collected
for all patients with suspected stroke for final adjudication and classification.
Any suspected stroke that occurred during the 30-day follow-up period was
independently adjudicated and confirmed by a central clinical events committee
(blinded to patient treatment assignment) and was classified as hemorrhagic
The rate of stroke and intracranial bleeding was determined among all
patients as randomized, and among all patients as treated, with exclusion
of the abciximab bolus-only group in the EPIC trial. Compared with the abciximab
bolus plus infusion treatment regimen, the abciximab bolus-only regimen in
the EPIC trial did not result in sufficient efficacy in reducing thrombotic
complications during PCI to warrant further clinical evaluation. Therefore,
all subsequent abciximab PCI trials used a treatment strategy that consisted
of a bolus followed by an infusion. For this reason, as well as to provide
an unbiased basis for comparison across the 4 trials and to reflect actual
clinical practice in which patients are being treated with abciximab bolus
and infusion, the abciximab bolus-only group was excluded from the treatment
analysis. Stroke rates in this group are reported separately.
Continuous variables are presented as means with SDs. Discrete variables
are shown as frequencies and percentages. Baseline characteristics among patients
with and without stroke were compared in univariable analysis using the t test for continuous variables and the Fisher exact test
for proportions. Statistical software used was S-PLUS, versioin 3.3 (Insightful
Corp, Seattle, Wash).
Among the 8555 patients randomized in the 4 trials, there were 33 strokes
reported in 31 patients (0.36%) within the first 30 days following enrollment.
Stroke occurred in 9 (0.29%) of the 3079 patients randomized to placebo and
in 22 (0.40%) of the 5476 patients randomized to abciximab (P = .46), including nonhemorrhagic stroke in 6 patients assigned placebo
(0.20%) and 13 patients assigned abciximab (0.24%) (P
= .81); hemorrhagic stroke or intracranial bleeding occurred in 3 patients
assigned placebo (0.10%) and in 9 patients assigned abciximab (0.16%) (P = .56). In 1 patient, the type of stroke could not be
Among the 695 patients in the EPIC abciximab bolus-only group, 1 patient
(0.14%) experienced a hemorrhagic stroke, and 4 patients (0.58%) had a nonhemorrhagic
Two strokes occurred among patients in the EPIC trial who had been randomized
to abciximab bolus plus infusion but were not treated because the stroke occurred
after randomization but before the angioplasty. Both patients (one with a
hemorrhagic stroke and one with a nonhemorrhagic stroke) died prior to study
drug administration and angioplasty. All patients who experienced strokes
or intracranial bleeding events in the CAPTURE, EPILOG, and EPISTENT trials
had received the allocated study drug.
Among the 7703 patients who received study drug bolus and infusion,
there were 24 strokes: 9 (0.30%) of 3023 patients treated with placebo and
15 (0.32%) of 4680 patients treated with abciximab bolus plus infusion. There
were no statistically significant differences in the rates between these groups
overall or in the rates of nonhemorrhagic stroke or intracranial bleeding/hemorrhagic
stroke (Table 1). One of 3 patients
with intracranial bleeding/hemorrhagic stroke in the placebo group died of
stroke compared with 4 of 7 patients treated with abciximab.
The rate of hemorrhagic stroke was higher in patients receiving abciximab
and standard-dose heparin (patients from EPIC, CAPTURE, and the standard-dose
heparin arm of EPILOG) compared with those receiving abciximab and low-dose
heparin (patients from EPISTENT and the low-dose heparin arm of EPILOG) (0.27%
vs 0.04%, respectively; P = .057) (Table 1). No differences were observed in the means of the maximum
activated clotting time values achieved during the procedure between patients
with hemorrhagic stroke and those without. A trend toward higher incidence
of nonhemorrhagic stroke was apparent in patients receiving abciximab and
low-dose heparin vs abciximab and standard-dose heparin (0.24% vs 0.09%, respectively; P = .30).
Patients who experienced a stroke were older, more often had a history
of hypertension, and less frequently had a history of diabetes mellitus (Table 2). Nonsignificant higher prevalences
of peripheral vascular disease and prior revascularization procedures were
apparent among patients who experienced a stroke.
This combined analysis of 8555 patients suggests that abciximab in addition
to aspirin and heparin does not increase the overall risk of stroke in patients
undergoing PCI. These results were obtained in a diverse population of patients
undergoing PCI, with or without stent deployment, varying from patients with
a low-risk profile and stable coronary artery disease to those at increased
risk during PCI because of unstable ischemic syndromes or unfavorable lesion
morphology on angiography.1- 4
For comparison, the rate of stroke in patients treated with heparin
undergoing PCI for stable coronary artery disease and unstable coronary syndromes,
including primary angioplasty for acute myocardial infarction, has been reported
to be between 0% and 1%.10- 19
In the current analysis, the rate of nonhemorrhagic stroke, as well as the
rate of intracranial hemorrhage, in patients treated with abciximab were comparable
to those for patients undergoing PCI receiving heparin.10- 19
The current data also are consistent with that of other intravenous GP IIb/IIIa
receptor inhibitors. In patients undergoing PCI in the IMPACT-II and RESTORE
trials, the incidence of intracranial hemorrhage associated with eptifibatide
and tirofiban treatment was 0.1%, which was comparable with the placebo event
rate.5,6 Similarly, in the 10 948
patients with acute coronary syndromes without persistent ST-segment elevation
in the PURSUIT trial, there were no differences in stroke rates between patients
who received placebo (0.8%, with hemorrhagic stroke rate <0.1%) and those
who received eptifibatide (0.6%, with hemorrhagic stroke rate <0.1%).20
In this study, patients who experienced a stroke were older, more often
had a history of hypertension, but less frequently a history of diabetes mellitus.
These findings are consistent with the established risk factors for both nonhemorrhagic
and hemorrhagic stroke among patients with ST-segment elevation myocardial
infarction treated with thrombolytic therapy,21- 25
and with the baseline clinical and demographic predictors of stroke in the
non–ST-segment elevation acute coronary syndrome patient population
in the PURSUIT trial.20
In patients treated with abciximab, a trend toward a higher incidence
of intracranial bleeding was observed among patients receiving standard-dose
heparin compared with those receiving a low-dose heparin regimen. As the number
of hemorrhagic strokes was small and the difference only reached borderline
statistical significance, definitive conclusions cannot be made about the
potential risk of heparin dosing on the incidence of hemorrhagic stroke in
the patient population studied. However, other major bleeding complications
are also reduced by the use of a low-dose heparin regimen compared with a
standard-dose heparin regimen in patients receiving abciximab.1- 4
Furthermore, previous studies in patients with acute myocardial infarction
treated with thrombolytic agents have shown that inappropriate high dosing
of antithrombotic therapy has the potential to substantially increase the
incidence of hemorrhagic stroke26- 28
and suggest that caution should be exercised with respect to dosing of heparin
in patients undergoing PCI and receiving abciximab. In fact, the EPILOG and
EPISTENT trials have shown that the clinical benefit of abciximab can be uncoupled
from the risk of hemorrhage by using low-dose, weight-adjusted heparin regimens.3,4 This approach should be regarded as
a standard treatment strategy for patients undergoing PCI and treated with
abciximab or any other GP IIb/IIIa receptor blocker.
On the other hand, surveillance with respect to the efficacy of the
low-dose heparin regimen in protecting patients from nonhemorrhagic stroke
is equally needed. Similarly, the apparent small excess of nonhemorrhagic
strokes in the abciximab bolus-only group in the EPIC trial compared with
the overall rate of nonhemorrhagic stroke among patients receiving abciximab
bolus plus infusion might also be an expression of the previously reported
insufficient level of antithrombotic efficacy provided by this treatment regimen
in this clinical setting.
The current analysis of 8555 patients undergoing PCI and treated with
either abciximab or placebo was limited by the very low incidence of stroke
in this patient population. As most strokes, and especially most hemorrhagic
strokes, occur in older patients, data on a larger number of elderly patients
are required to provide more definitive assurance of the lack of risk of stroke
in this subpopulation. Continued surveillance for occurrence of stroke in
patients treated with abciximab and systemic collection of follow-up data
from clinical trials and in clinical practice are needed.