CREATE indicates Clinical Trial of Reviparin and Metabolic Modulation
in Acute Myocardial Infarction Treatment Evaluation; ECLA, Estudios Cardiologicas
Latin America study group; GIK, glucose, insulin, and potassium; UFH, unfractionated
heparin; LMWH, low-molecular-weight heparin.
HR indicates hazard ratio; CI, confidence interval. There is an early
separation of the curves with continuing divergence over the entire 7-day
HR indicates hazard ratio; CI, confidence interval. There were 52 reinfarctions
within 24 hours in the placebo group and 40 reinfarctions within 24 hours
in the reviparin group.
There is an increasing benefit with earlier treatment (P = .01 for trend) but the effects in those patients with
or without reperfusion therapy did not significantly differ from each other
(P = .31 for interaction). The boxes indicate
hazard ratios, and their size is proportional to the number of patients in
HR indicates hazard ratio; CI, confidence interval. There is a substantially
greater benefit from reviparin in preventing major vascular events (18 fewer
per 1000 patients) compared with the small excess in other life-threatening
bleeding (1 more per 1000 patients).
Customize your JAMA Network experience by selecting one or more topics from the list below.
The CREATE Trial Group Investigators*. Effects of Reviparin, a Low-Molecular-Weight Heparin, on Mortality, Reinfarction, and Strokes in Patients With Acute Myocardial Infarction Presenting With ST-Segment Elevation. JAMA. 2005;293(4):427–435. doi:10.1001/jama.293.4.427
Context Although reperfusion therapy, aspirin, β-blockers, and angiotensin-converting
enzyme inhibitors reduce mortality when used early in patients with acute
myocardial infarction (MI), mortality and morbidity remain high. No antithrombotic
or newer antiplatelet drug has been shown to reduce mortality in acute MI.
Objective To evaluate the effects of reviparin, a low-molecular-weight heparin,
when initiated early and given for 7 days in addition to usual therapy on
the primary composite outcome of death, myocardial reinfarction, or strokes
at 7 and 30 days.
Design, Setting, and Patients A randomized, double-blind, placebo-controlled trial (Clinical Trial
of Reviparin and Metabolic Modulation in Acute Myocardial Infarction Treatment
Evaluation [CREATE]) of 15 570 patients with ST-segment elevation or
new left bundle-branch block, presenting within 12 hours of symptom onset
at 341 hospitals in India and China from July 2001 through July 2004.
Intervention Reviparin or placebo subcutaneously twice daily for 7 days.
Main Outcome Measure Primary composite outcome of death, myocardial reinfarction, or stroke
at 7 and 30 days.
Results The primary composite outcome was significantly reduced from 854 (11.0%)
of 7790 patients in the placebo group to 745 (9.6%) of 7780 in the reviparin
group (hazard ratio [HR], 0.87; 95% CI, 0.79-0.96; P = .005).
These benefits persisted at 30 days (1056 [13.6%] vs 921 [11.8%] patients;
HR, 0.87; 95% CI, 0.79-0.95; P = .001)
with significant reductions in 30-day mortality (877 [11.3%] vs 766 [9.8%];
HR, 0.87; 95% CI, 0.79-0.96; P = .005)
and reinfarction (199 [2.6%] vs 154 [2.0%]; HR, 0.77; 95% CI, 0.62-0.95; P = .01), and no significant differences in strokes
(64 [0.8%] vs 80 [1.0%]; P = .19). Reviparin
treatment was significantly better when it was initiated very early after
symptom onset at 7 days (<2 hours: HR, 0.70; 95% CI, 0.52-0.96; P = .03; 30/1000 events prevented; 2 to <4 hours: HR,
0.81; 95% CI, 0.67-0.98; P = .03; 21/1000
events prevented; 4 to <8 hours: HR, 0.85; 95% CI, 0.73-0.99; P = .05; 16/1000 events prevented; and ≥8 hours: HR, 1.06;
95% CI, 0.86-1.30; P = .58; P = .04 for trend). There was an increase in life-threatening
bleeding at 7 days with reviparin and placebo (17 [0.2%] vs 7 [0.1%], respectively; P = .07), but the absolute excess was small (1
more per 1000) vs reductions in the primary outcome (18 fewer per 1000) or
mortality (15 fewer per 1000).
Conclusions In patients with acute ST-segment elevation or new left bundle-branch
block MI, reviparin reduces mortality and reinfarction, without a substantive
increase in overall stroke rates. There is a small absolute excess of life-threatening
bleeding but the benefits outweigh the risks.
Approximately 15.5 million cardiovascular deaths occur every year.1 Of these, about half are likely to be due to acute
myocardial infarction (MI), with the majority occurring in low- and middle-income
countries. Aspirin,2 thrombolytic therapy,3 β-blockers,4 and
angiotensin-converting enzyme (ACE) inhibitors5 improve
prognosis in acute MI. Primary percutaneous coronary angioplasty (PCI) offers
benefits over thrombolytic therapy,6 but access
to primary PCI is limited and not affordable to the majority of patients in
Combinations of newer antiplatelet regimens,7,8 or
direct thrombin inhibitors,9,10 appear
to reduce reinfarction but do not reduce mortality.7-10 Moreover,
these agents are expensive and increase bleeding. Although intravenous unfractionated
heparin is commonly used after acute MI, especially in patients receiving
fibrin-specific thrombolytic agents, this practice is based on a few trials
that indicated modest improvements in coronary patency.11-14 However,
no reduction in mortality or reinfarction has been documented and there appears
to be an increase in bleeding. In the Third International Study of Infarct
Survival (ISIS-3)15 and Gruppo Italiano per
lo Studio della Sopravvivenza nell’Infarto Miocardico (GISSI-2)16 trial groups, subcutaneous heparin reduced in-hospital
mortality and reinfarction to a small extent, when given with fibrin-specific
and nonspecific thrombolytic agents, but these differences did not persist
at 1 month. In a subsequent large trial,17 intravenous
heparin tended to be worse than subcutaneous heparin, when given with streptokinase.
Trials of enoxaparin (vs placebo or unfractionated heparin) indicate fewer
reinfarctions,8 but none observed mortality
reductions,18 and there are concerns of substantial
increases in strokes when enoxaparin and some fibrin-specific agents such
as tenectoplase are used in elderly patients.19
Therefore, the net benefit-risk ratio of using any thrombin inhibitor
in acute MI is not clear. A meta-analysis of the trials comparing enoxaparin
vs unfractionated heparin suggests a lower rate of reinfarction, with no impact
on mortality but an increase in major bleeding and strokes.18 Given
that unfractionated heparin does not reduce mortality or reinfarction (vs
usual care) but increases bleeding, trials of low-molecular-weight heparin
(LMWH) vs unfractionated heparin are difficult to interpret for both efficacy
or safety. However, given the promising data with LMWH in patients with ST-segment
elevation acute MI, a large definitive trial is urgently needed.
We conducted 2 similar trials, the Clinical Trial of Reviparin and Metabolic
Modulation in Acute Myocardial Infarction Treatment Evaluation (CREATE) trial
and the Estudios Cardiologicas Latin America Study Group trial,18 aimed
at reliably evaluating the role of high-dose glucose, insulin, and potassium
(GIK) in 20 000 patients and also described in the accompanying article.20 In the 2 largest countries (India and China), we
simultaneously evaluated reviparin (an LMWH) in 15 570 patients (CREATE
trial) using a factorial design. The goal of this part of the trial was to
evaluate the impact of reviparin on the composite outcome of death, myocardial
reinfarction, and stroke (first coprimary outcome) or the first coprimary
outcome with the addition of recurrent ischemia with electrocardiogram changes
(second coprimary outcome) at 7 and 30 days. The impact on individual components
of the composite outcomes, especially mortality, were evaluated. A combined
efficacy and safety outcome of death, myocardial reinfarction, strokes, and
life-threatening bleeding at 30 days was used to assess the balance between
efficacy and safety.
The CREATE trial used a partial 2 × 2 factorial design
comparing reviparin with placebo administered for 7 days (double-blind) and
comparing GIK with control (open) administered for 24 hours, in addition to
usual care. Patients in India and China were randomized to both parts of the
trial (n = 15 570 patients), whereas those from other countries
were only included in the GIK component. All centers obtained local ethics
committee approval and, in addition, the project office obtained approval
from the institutional review board of the Hamilton Health Sciences and McMaster
University, Hamilton, Ontario.
Following written or witnessed oral informed consent, patients presenting
with suspected acute MI and ST-segment elevation or new left bundle-branch
block within 12 hours of symptom onset and without contraindications to heparin
(active or high risk of bleeding, recent major surgery or trauma within 2
weeks, systolic blood pressure ≥180 mm Hg, severe anemia, hemorrhagic stroke
<12 months, oral anticoagulant therapy, heparin-induced thrombocytopenia,
pregnancy, or other diseases limiting life expectancy less than 1 month) or
GIK (type 1 diabetes mellitus, renal impairment, or hyperkalemia) were randomized
Study drugs were recommended to be initiated prior to or within 15 minutes
of thrombolytic therapy. Patients who weighed less than 50 kg received 3436
IU Ph Eur antiXa units of reviparin (provided by Abbott GmbH & Company
KG, Ludwigshafen, Germany) every 12 hours subcutaneously, patients who weighed
between 50 to 75 kg received 5153 IU every 12 hours, and patients who weighed
more than 75 kg received 6871 IU every 12 hours. In patients undergoing primary
PCI, open-label unfractionated heparin was used during the procedure, with
study medication being initiated 1 hour after removal of the sheath. Nonstudy
thrombin inhibitors were not allowed, unless there was a clinical need, in
which case patients who were blinded to study medication were discontinued
of study drug. All other proven therapies were permitted based on the physician’s
Data from 274 centers in China and 67 centers in India were sent to
the National Coordinating Office (NCO) at the Beijing Hypertension League
Offices, Beijing, China, or to the NCO at St John’s Academy of Medical
Sciences, Bangalore, India, respectively. Randomization to reviparin or placebo
was grouped in blocks, with the block size kept confidential and the randomization
list stratified by center. All patients in China were randomized by telephone
to Beijing. In India, patients were initially randomized using sealed opaque
envelopes (n = 5127), but this procedure was changed to central
telephone randomization for subsequent patients (n = 2933). Despite
extensive precautions, randomization errors occurred in 173 (1.1%) of 15 570
patients. These patients were included in the originally intended allocation
The NCOs entered the data into a Web-based database that was connected
online to the Population Health Research Institute, McMaster University and
Hamilton Health Sciences, Hamilton, Ontario. Extensive consistency and edit
checks at the NCO and the Population Health Research Institute, and at site
visits ensured high data completeness (99.9%).
Follow-up at 30 days was obtained by having patients return for a no-cost
visit. For approximately 200 patients in which follow-up was difficult to
obtain, even after telephone calls and mailings, research staff visited the
patients’ homes. Patients were not paid to participate in the trial
but were offered transportation costs and, as noted, a no-cost 30-day follow-up
visit. Centers were reimbursed based on each follow-up visit completed.
The first coprimary outcome was a composite of death, reinfarction,
or stroke at 7 days, while the second coprimary outcome was the first outcome
with the addition of ischemia with electrocardiogram changes, also at 7 days.
Secondary outcomes included the components of these coprimary outcomes, any
ischemia at 7 days, and 30-day outcomes, which included the same composites.
Total mortality was included in the composite efficacy outcomes. Causes
of death were categorized by the local investigator without knowledge of treatment
allocation. Reinfarction was defined as recurrent typical chest pain with
new clear persistent ischemic (ST reelevation or depression) electrocardiogram
changes within 24 hours; or after 24 hours, recurrent typical chest pain with
characteristic new electrocardiogram changes (new Q waves, ST elevation or
depression with evolution) or a further increase in enzyme levels (to twice
the upper limit of normal if it had returned to baseline or if already elevated,
with a further elevation by 50%). Refractory ischemia was defined as recurrent
chest pain on optimal medical therapy (at least 2 anti-anginals) and further
subcategorized as those patients with or without documented new electrocardiogram
changes. Strokes were defined as focal neurological deficits that persisted
for more than 24 hours. These were further categorized as probable/definite
hemorrhagic, ischemic, or type unknown. Computed tomographic scans, magnetic
resonance imaging, or autopsy were available in 102 (70.8%) of 144 patients
with strokes. Major bleeding was defined as bleeding requiring the transfusion
of at least 2 units of blood, intracranial or fatal. Major bleeding was subcategorized
as life-threatening (fatal, needing surgical intervention, transfusion of ≥4
units of blood, drop in hemoglobin of >5 g/dL, or intracranial) or non–life-threatening.
Intracranial bleeding and fatal bleeding were already included in the efficacy
composite outcomes. All efficacy and safety outcomes were adjudicated by blinded
central committees in each country.
An independent data and safety monitoring board regularly reviewed the
accumulating data. Three formal interim analyses occurred when 25%, 50%, and
75% of the data were available. For the first 2 interim analyses, the boundary
for benefit had to exceed a difference of 4 SDs (χ2 = 16; P<.0001) for the first coprimary outcome for the reviparin
group, on 2 successive examinations of the data about 3 months apart. For
the third analysis, the boundary was 3.5 SDs (χ2 = 12.3; P<.00047).
Anticipating a 12% rate for the first coprimary outcome of death, reinfarction,
or stroke at 7 days in the placebo group with 15 000 patients, there
was 93% power to detect a 15% relative risk reduction with reviparin.
All analyses presented are by intention-to-treat. Where randomization
errors occurred (1.1%), analyses are based on the group they were originally
allocated to rather than the treatment actually administered. This resulted
in the opposite treatment being administered to 0.55%; our analytical approach
led to a very slight underestimation of the treatment effect. The study was
considered significant at P≤.05 for both coprimary
outcomes, or if the first coprimary outcome was P≤.046
or the second coprimary was P≤.01, which preserved
the overall α level at .05. Outcomes were compared using hazard ratios
(HRs) and the 95% confidence intervals (CIs) derived from a Cox proportional
hazards regression model; the proportional hazards assumption was confirmed.
Data were also presented using Kaplan-Meier curves. Practically no adjustments
of the final P values were required given the extreme
statistical boundaries used for interim analyses. Subgroup hypotheses were
analyzed using tests for trend or interactions in the Cox proportional hazards
regression analyses. Two prespecified subgroup hypotheses were postulated:
first, that treatment would be effective both in those patients undergoing
and not undergoing reperfusion therapy; and second, that reviparin would be
more effective with earlier treatment. Statistical analyses were performed
with SAS version 8.2 (SAS Institute Inc, Cary, NC).
A total of 15 570 patients were randomized to receive either reviparin
or placebo. Seven-day data were available in all patients (100%) and 30-day
vital status was known in 15 565 patients (99.96%). A total of 7780 patients
received reviparin and 7790 received placebo.
Table 1 summarizes key baseline
characteristics. More than 75% of the patients were randomized within 8 hours
of symptom onset (median duration, 4.9 hours). A total of 13 030 patients
(83.7%) were in Killip class I (no pulmonary rales or third heart sound).
Thrombolytic therapy was used in 11 355 patients (73%) and primary PCI
in 949 patients (6.1%), with any reperfusion strategy in 12 245 patients
(79%) (59 patients received both). Aspirin was used in 15 084 patients
(96.9%), clopidogrel or ticlopidine in 8555 (54.9%), β-blockers in 10 259
(65.9%), ACE inhibitors in 11 314 (72.7%), and lipid-lowering medications
in 10 374 (66.6%). The number of patients that evolved electrocardiographic
Q-waves was 12 064 (77.5%) overall, with no significant difference between
the 2 groups. The number of patients undergoing PCI after thrombolytic therapy
was 156 (2.7%) in the reviparin group compared with 201 (3.5%) in the placebo
group (P = .02).
A total of 7626 patients (98.0%) allocated to reviparin and 7647 (98.2%)
allocated to placebo received study drugs (Figure
1). The time from randomization to administration of the first dose
of study drug was less than 1 hour in 85% of patients. In 2745 patients (17.6%),
the study drug was initiated before thrombolytic therapy; in 8512 patients
(54.7%), the study drug was administered after thrombolytic therapy, with
about a third (n = 2750) receiving it less than 30 minutes, another
third within 30 to 60 minutes (n = 2294), and another third more
than 60 minutes (n = 3468). Prerandomization heparin was used in
1466 patients (9.4%) and nonstudy heparin or LMWH after randomization was
used in 739 patients (9.5%) in the reviparin group compared with 799 (10.3%)
in the placebo group. A total of 11 893 patients (76%) received allocated
study drug for 7 days, with 14 231 (91.4%) receiving it for at least
Both coprimary outcomes were significantly reduced at 7 days with reviparin
(Table 2). The composite efficacy of
death, myocardial reinfarction, or strokes was reduced from 854 patients (11.0%)
in the placebo group to 745 (9.6%) in the reviparin group (HR, 0.87; 95% CI,
0.79-0.96; P = .005) (Figure 2). The second coprimary outcome, which included recurrent
ischemia with electrocardiogram changes, was also significantly reduced from
982 patients (12.6%) to 864 (11.1%) (HR, 0.87; 95% CI, 0.80-0.96; P = .004). Significant reductions in mortality and myocardial
reinfarction were also observed, with favorable trends toward lower rates
of recurrent ischemia. However, there was no significant difference in strokes.
There was a mix of a small but significant excess of hemorrhagic strokes (23
patients [0.3%] vs 10 [0.1%]; P = .03),
with little difference in other types of strokes (38 patients [0.5%] vs 39
At 30 days, both composite outcomes were similarly reduced (Table 3). The reductions in mortality and myocardial
reinfarction were highly significant (Figure 3), with no significant excess in strokes (Figure 4). The number of patients with a composite outcome of death,
reinfarction, or disabling strokes at 30 days were signficantly reduced from
1034 patients (13.3%) in the placebo group to 904 (11.6%) in the reviparin
group (HR, 0.87; 95% CI, 0.79-0.95; P = .002).
The benefits of reviparin were greatest with earlier treatment after
symptom onset at 7 days (<2 hours: HR, 0.70; 95% CI, 0.52-0.96; P = .03; 30 events prevented per 1000 patients; 2 to <4
hours: HR, 0.81; 95% CI, 0.67-0.98; P = .03;
21 events prevented per 1000 patients; 4 to <8 hours: HR, 0.85; 95% CI,
0.73-0.99; P = .05; 16 events prevented
per 1000 patients; and ≥8 hours: HR, 1.06; 95% CI, 0.86-1.30; P = .58; P = .04
for trend). Similarly, at 30 days, a 30% relative risk reduction was observed
in those patients randomized less than 2 hours compared with 20% in those
randomized between 2 to 4 hours, 15% between 4 and 8 hours, and little benefit
for those randomized more than 8 hours (P = .01
for trend) (Figure 5). Similar trends
were observed for mortality and myocardial reinfarction but not for strokes.
Consistent benefits at 7 days were observed in those patients undergoing reperfusion
therapy (HR, 0.90; 95% CI, 0.81-1.01) and in those not receiving this therapy
(HR, 0.79; 95% CI, 0.65-0.95; P = .23 for
interaction) for the first coprimary outcome, with similar results for the
second coprimary outcome (HR, 0.90; 95% CI, 0.81-1.00; and HR, 0.81; 95% CI,
0.68-0.97; respectively). Similar results with benefits in both subgroups
were also observed by 30 days. In the subgroup of patients (n = 949)
undergoing primary PCI, trends toward fewer events were observed (28 [5.8%]
of 481 patients vs 34 [7.3%] of 468; HR, 0.79; 95% CI, 0.48-1.31; for the
first coprimary outcome, and 35 [7.3%] of 481 vs 47 [10.0%] of 468; HR, 0.71;
95% CI, 0.46-1.10; for the second coprimary outcome). In the patients who
received tissue plasminogen activator or primary PCI (which approximates the
usual practice in many North American centers), there were 31 (5.8%) of 532
individuals assigned to reviparin with death, myocardial reinfarction, stroke,
or recurrent ischemia compared with 40 (7.7%) of 520 individuals assigned
to placebo (HR, 0.75; 95% CI, 0.47-1.19), suggesting that the benefits of
reviparin were independent of the type of reperfusion therapy.
There was a significant increase in the rates of life-threatening or
major bleeding at 7 days (Table 4).
The most common sites were gastrointestinal (19 vs 9 patients) and intracranial
(22 vs 10 patients). The increased bleeding risk tended to be greater in those
patients undergoing reperfusion therapy (1.1% receiving reviparin vs 0.4%
receiving placebo), whereas the rates were low in the 3325 patients without
reperfusion therapy (0.1% vs 0.1%, respectively). Intracranial hemorrhage
also tended to be higher with reviparin in those patients undergoing reperfusion
therapy (0.4% vs 0.1%) compared with those not undergoing such treatment (0.1%
vs 0.1%). However, fatal bleeding and intracranial bleeding were already included
in the analysis of death or strokes, as part of the efficacy outcome. Therefore,
the number of patients with life-threatening bleeding not associated with
death or strokes was 17 (0.2%) receiving reviparin vs 7 (0.1%) receiving placebo
(P = .07).
The composite outcome of death, myocardial reinfarction, strokes, and
life-threatening bleeding (Figure 6)
occurred in 762 patients (9.8%) in the reviparin group compared with 861 (11.1%)
in the placebo group (HR, 0.88; 95% CI, 0.80-0.97; P = .01),
with similar results at 30 days (934 [12.0%] vs 1065 [13.7%] patients; HR,
0.87; 95% CI, 0.80-0.95; P = .002), which
suggests that for every 1000 patients treated with reviparin, 17 fewer major
adverse outcomes would be prevented.
The CREATE trial demonstrated a clear and significant reduction in both
coprimary outcomes at 7 and 30 days. Furthermore, there were significant reductions
in death and myocardial reinfarction, with no significant excess in strokes.
There was a small but significant early increase in life-threatening bleeding,
but this only slightly offset the reductions in mortality or myocardial reinfarction.
Overall, there were 17 fewer major events (death from any cause, reinfarction,
strokes, or life-threatening bleeding) per 1000 patients who were administered
reviparin (P = .002). Thus, the overall
benefits of reviparin in acute MI clearly outweighs its risks.
Reductions in recurrent myocardial reinfarction and recurrent ischemic
events observed in the CREATE trial with reviparin are consistent with the
results of previous trials of LMWH, such as the ASSENT-3 study,8,15 which
used enoxaparin, or those studies evaluating direct thrombin inhibitors.9,10 However, unlike these trials, the
CREATE trial has also observed significant reductions in mortality. Although
there was an excess of early intracranial bleeding, there was no significant
excess of overall strokes with the absolute excess being small (2 per 1000
patients treated). Eighty percent of patients with intracranial hemorrhagic
died; therefore, these deaths due to strokes were included in the analysis
of deaths, which was still significantly lower with reviparin. The rates of
intracranial hemorrhage and strokes in our trial (0.3% and 0.8%, respectively)
in the reviparin group at 7 days were lower than previous trials of enoxaparin
used in conjunction with fibrin-specific thrombolytic agents, such as tenecteplase
(for intracranial hemorrhage and total strokes, 0.88% and 1.62%, respectively,
in the ASSENT-3 study8; and 2.2% and 2.9%,
respectively, in the ASSENT-Plus study14).
Therefore, given the clearly lower mortality and myocardial reinfarction rates
with no excess in overall stroke rates, the combination of reviparin with
less expensive nonspecific thrombolytic therapy is a reasonable option in
patients with acute MI receiving other effective therapies.
Previous trials of unfractionated heparin in patients with acute ST
elevation MI who are receiving streptokinase or tissue plasminogen activator
and aspirin have not demonstrated a reduction in major vascular events at
1 month, despite increased bleeding.15,16 In
these trials, there was a small reduction in hospital mortality and reinfarction,
but these differences did not persist at 1 month. This lack of benefit may
have been because of the inherent limitations of unfractionated heparin or
delays in therapy after a thrombolytic agent (>4 hours in ISIS-315 and
>12 hours in GISSI-216). In the GUSTO trial,17 there was no significant impact on mortality but
an increase in reinfarction with intravenous heparin compared with subcutaneous
heparin when used in conjunction with streptokinase.13 Intravenous
heparin is commonly used in some countries for approximately 24 to 48 hours
after a fibrin-specific thrombolytic agent, largely based on the results of
some but not all trials that demonstrated differences in angiographic patency.11-14 However,
this difference in patency was very small in the only trial that used an adequate
dose of aspirin.14 Even collectively, these
trials were too small to reliably evaluate the impact on mortality, reinfarction,
or strokes, but was associated with an increase in major bleeding in 1 study.21 Thus, despite the common use of intravenous heparin
with a fibrin-specific thrombolytic agent, the net benefit-risk ratio on clinical
outcomes is uncertain.
Low-molecular-weight heparin has been evaluated in ST elevation acute
MI in several small- and moderate-sized studies. No trial has demonstrated
a significant reduction in mortality, perhaps because of the small size, shorter
duration of treatment, and higher rates of intracranial bleeding. A meta-analysis
of 6 randomized controlled trials of enoxaparin vs unfractionated heparin
in approximately 8000 patients indicates a reduction in myocardial reinfarction
with enoxaparin (3.2% vs 5.1%, respectively; odds ratio [OR], 0.61; 95% CI,
0.48-0.76), no impact on mortality (5.8% vs 6.1%, respectively; OR, 0.97;
95% CI, 0.81-1.17), an increase in major bleeding (3.2% vs 2.3%, respectively;
OR, 1.38; 95% CI, 1.05-1.81), and hemorrhagic strokes (1.29% vs 0.89%, respectively;
OR, 1.30; 95% CI, 0.84-2.03)18 compared with
Prior data comparing LMWH with placebo are also sparse (only 1376 patients)
and indicate little impact on mortality (6.4% vs 6.8%, respectively; OR, 0.75;
95% CI, 0.36-1.55).18 However, the CIs were
wide and are consistent with the CREATE trial results. In these trials, there
was a significant reduction in reinfarction (3.2% with LMWH vs 6.0% with placebo;
OR, 0.54; 95% CI, 0.33-0.91), but much higher rates of major bleeding (3.6%
vs 1.0%, respectively; OR, 3.0; 95% CI, 1.50-6.00). The results of the CREATE
trial, which involved 4 times the number of patients compared with any previous
trials of LMWH, demonstrates significant reductions in death and myocardial
reinfarction. Although there was increased bleeding, this adverse event was
much smaller than that observed with trials of other LMWHs, newer antiplatelet
drugs, or direct thrombin inhibitors.7-10,18
Interestingly, the reductions in death and myocardial reinfarction at
7 and 30 days were greater with earlier treatment but no such trend was observed
with strokes alone. This suggests that at least some of the benefits of reviparin
may be related to improved rates of coronary patency and increased myocardial
salvage. Because the benefits are related to the time of initiation of treatment
and the adverse effects are not, the benefit-risk balance is likely to be
most favorable if treatment is initiated with 8 hours or even earlier. In
the CREATE trial, administering 1000 patients with reviparin who presented
8 hours or less at the hospital for 7 days prevented 20 major vascular events,
at a cost of only 2 other life-threatening bleeding.
Reperfusion therapy was administered in 78.6% of patients in our trial
and benefits were noted among those patients receiving or not receiving such
therapy. In particular, there was an apparent consistent benefit in those
patients undergoing primary PCI, suggesting that prolonged antithrombin therapy
after the procedure may well be beneficial.
The CREATE trial was conducted in India and China, the 2 largest countries
in the world, which are projected to have approximately 50% of the global
burden of coronary heart disease. Our results may be applicable to other countries
for several reasons. First, the rates of use of proven pharmacological treatments,
such as aspirin (97%), β-blockers (66%), ACE inhibitors (73%), lipid-lowering
therapy (67%), and reperfusion therapy (79%), were high. Second, although
streptokinase and urokinase were the thrombolytic agents most commonly used,
there was a consistent trend toward benefit in the 1052 patients undergoing
primary PCI or receiving tissue plasminogen activator (HR, 0.65; 95% CI, 0.37-1.13).
Furthermore, there was a clear benefit among those patients not receiving
reperfusion therapy (HR, 0.79; 95% CI, 0.65-0.95). Therefore, reviparin was
effective irrespective of the ancillary treatments used. Third, the duration
of hospitalization may be shorter by a day or 2 in some Western countries,
but the benefits in the CREATE trial emerged early and patients can self-inject
reviparin after hospital discharge, as has been performed in deep vein thrombosis
prophylaxis. Therefore, use of an LMWH, such as reviparin, for a week should
be practical and effective in most settings. Finally, the high adherence to
the protocol and very high rates of complete follow-up provides high confidence
in the validity and broad applicability of our findings.
In common with all trials of antithrombotic and thrombolytic therapy,
we observed a significant increase in bleeding rates for those events classified
as either life-threatening or other major bleeding. We deliberately chose
our primary outcome to include total mortality, myocardial reinfarction, and
stroke, because these outcomes include intracranial bleeding and other fatal
bleeding, and provide an estimate of the balance between benefit and risks.
A further analysis of the outcome of death, reinfarction, strokes, and life-threatening
bleeding at 30 days indicates a significant risk reduction of 0.87 (95% CI,
0.80-0.95; P = .002), suggesting a moderate
overall benefit with reviparin when added to conventional therapies. Treatment
of 1000 patients for 7 days prevented 18 deaths or reinfarctions, with an
excess of 1 other life-threatening bleeding (net benefit of 17 events per
Reviparin has been previously shown to be beneficial in preventing deep
vein thrombosis.22 It has a relatively low
molecular weight (3900 Da) and high anti-Xa activity.23 The
anti-Xa/IIa ratio is about 3.3, which is similar to enoxaparin (3.3) and nadroparin
(3.0), but higher than that of dalteparin (2.0) and tinzaparin (1.8).23 However, given that dalteparin and enoxaparin are
beneficial in patients with non–ST-elevation MI and unstable angina,24,25 and the data with enoxaparin in ST-elevation
MI are promising, it is reasonable to expect some benefits from other LMWHs.
However, the exact dose of these agents and their benefits and bleeding risks,
especially in the context of thrombolytic agents requires clarification. Therefore,
further evaluation for each agent at specific doses is required, before these
agents can be widely used in the treatment of acute MI. At least 1 such large
trial (Enoxaparin and Thrombolysis Reperfusion for Acute Myocardial Infarction
Treatment Thrombolysis in Myocardial Infarction–Study 25) evaluating
enoxaparin is under way and should provide complementary information.26 The increased risk of bleeding with various antithrombin
therapies (heparin, LMWH, direct thrombin inhibitors), intravenous glycoprotein
IIb/IIIa inhibitors as well as new thrombolytic agents emphasizes the need
to evaluate both clinical benefits and risks in large trials, rather than
reliance on surrogate end points, before these drugs are used in clinical
The incremental benefits of reviparin in patients with ST-segment elevation
were moderate but clinically worthwhile. It is of the order of magnitude that
has been claimed with accelerated tissue plasminogen activator vs streptokinase
but with no excess of overall strokes.13,22 Moreover,
the benefits of reviparin has been more clearly demonstrated than with any
other antithrombotic therapy (hirudin or bivalirudin) or intravenous glycoprotein
IIb/IIIa inhibitors in the treatment of acute MI. In these trials, the duration
of treatment was relatively short (48-96 hours), whereas in the CREATE trial,
it was 7 days. There was excellent adherence to our protocol with 61.5% of
patients receiving treatment within 6 hours along with thrombolytic agents
and continuing reviparin therapy for the full duration of the trial. These
factors, the high rates of follow-up and data quality, and the large size
of our study likely contributed to the clear results of the CREATE trial.
Reviparin is considerably less expensive than other antithrombotic agents,
such as bivalirudin, is somewhat cheaper compared with other LMWHs, and can
be given subcutaneously. Its use is relatively straightforward and can be
used in both developed and developing countries. Therefore, the benefits of
reviparin represents a moderate but important globally applicable advance
in the management of patients with acute MI.
Corresponding Author: Salim Yusuf, DPhil,
FRCPC, Population Health Research Institute, Hamilton General Hospital and
McMaster University, 237 Barton St E, Hamilton, Ontario, Canada L8L 2X2 (email@example.com).
Author Contributions: Dr Yusuf had full access
to all of the data in the study and takes responsibility for the integrity
of the data and the accuracy of the data analysis.
Study concept and design: Yusuf, Mehta, Xavier,
Pais, Zhu, Liu.
Acquisition of data: Yusuf, Mehta, Xie, Ahmed,
Xavier, Pais, Zhu, Liu.
Analysis and interpretation of data: Yusuf,
Mehta, Xavier, Pais, Zhu, Liu.
Drafting of the manuscript: Yusuf, Mehta.
Critical revision of the manuscript for important
intellectual content: Yusuf, Mehta, Xie, Ahmed, Xavier, Pais, Zhu,
Statistical analysis: Xie.
Obtained funding: Yusuf.
Administrative, technical, or material support:
Study supervision: Yusuf, Mehta.
List of Centers and Investigators: The following
persons participated in the CREATE Study.
International Steering Committee: S. Yusuf
(principal investigator and chairman), S. R. Mehta (coprincipal investigator
and project director), P. Pais (principal investigator, India), S. Reddy (coprincipal
investigator, India), L. Liu (principal investigator, China), R. J. Ahmed,
L. Cronin (global study coordinators), D. Xavier (coordinator, India), J.
Zhu (coordinator, China), C. Xie (project statistician); Indian Steering Committee: R. Gupta, K.K. Haridas, T. M. Jaison, P.
P. Joshi, P.G. Kerkar, A.K. Maity, S.C. Manchanda, S. Naik, P. Pais, D. Prabhakaran,
S. Reddy, B. Singh, S. Thanikachalam, D. Xavier; Chinese
Steering Committee: X. J. Bai, T. Chen, J. J. Cui, T. X. Cui, S. Y.
Fu, H. Ge, Q. L. Li, S. M. Li, W. Li, Y. Q. Li, L. Liu, Y. H. Liu, Z. R. Lu,
S. P. Ma, D. Qiao, Y. C. Song, N. L. Sun, L. H. Wang, S. W. Wang, W. Wang,
Y. Wang, N. Wu, Y. S. Wu, C. B. Xu, S. C. Xu, Z. M. Xu, G. J. Yang, H. S.
Yang, C. Z. Zhang, S. T. Zhang, W. J. Zhang, J. C. Zhou, J. Zhu; Population Health Research Institute, Hamilton, Canada: S. Yusuf, S.
R. Mehta, R. J. Ahmed, L. Cronin, S. Pavlov, C. Xie, J. Pogue, F. Zhao, I.
Tsuluca, M. Molec, I. Holadyk-Gris, K. Ahmed; Indian National
Coordinating Office, Institute of Population Health and Clinical Research,
St John's National Academy of Health Sciences, Bangalore, India: P.
Pais, D. Xavier, D. Freeda, S. Lidwin; Indian Adjudication
Committee: M. Chenniappan, B. Isaac, S. S. Iyengar, T. M. Jaison, P.
Joshi, S. P. Kalantri, S. K. Kaushik, P. G. Kerkar, U. K. Mahorkar, J. Narendra,
S. K. Paul, M. J. Santhosh, B. K. S. Sastry, B. Singh, S. B. Siwach, K. Varghese; China National Coordinating Office, Beijing Hypertensive League
Institute, Beijing, China: L. Liu, J. Zhu, H. Yang, Y. Yang, X. Zhang,
H. Tan, J. Tang, X. Li, L. Yan, Y. Zhang, J. Li; Chinese
Adjudication Committee: M.Y. Bai, Y. Q. Jiang, S. Y. Lang, X. Y. Shi,
Y. C. Song, Z. R. Tian, K. Wang, D. H. Yan, S. Y. Yu; Data
and Safety Monitoring Board: P. Sleight (Chair), C. Baigent, J. Hirsh,
W. Taylor, G. Tognoni; Abbott GmbH & Company KG (reviparin
sponsor): P. Bacher, N. Bender, U. Legler, U. Magin, U. Raschke; China:An Yang: H. Liu; Angang: R. Wang; Anshan: Z. C.
Liu, X. Tian, G. Wang, Y. Zhang; Baoding: Z. Nan,
J. Zhang; Bazhou: C. Zai; Beijing: W. Chen, M. Gao, D. Hu, S. Jia, D. X. Li, Q. Li, W. Li, S. L. Liu,
Y. Sun, B. Wang, G. F. Xie, Z. Xu, X. Yang, M. Zhao, X. Zhao; Beipiao: Z. Fang; Benxi: S. Y. Liu; Cangzhou: Z. Ma; Changchun: Y.
Jiang, S. M. Li; Changsha: Z. Zhen; Chendu: F. Huo; Chengde: H. G. Yang; Chengwu: H. Liu; Chifeng: Y. Miao; Chongqing: C. M. Yang; Dalian:
X. Chi, Z. X. Liu, S. Zhou; Dandong: Y. Sun; Dashiqiao: F. S. Zhou; Fenyang:
R. Guo; Gaoping: K. Jing; Guan: Z. Xu; Haicheng: S. Ren, J. Zhao; Hebei: H. Bai, H. Bai, C. Cheng, J. Cheng, X. Hao, H. Li,
W. G. Li, S. Wang, W. Zhang; Heilongjiang: L. Li,
Y. Sun; Helongjiang: S. Fu, J. Shao, X. Tan; Henan: S. Chen, J. Fu; Hengshui:
Q. Zheng; Huai Ren: J. Ma; Inner
Mongolia: H. Chen, H. Ma; Jiamusi: L. Gong; Jiaozhou: Z. Zhang; Jiaxiang: F.
Li; Jilin B. Yang; Jinan: L.
S. Zhou; Jinlin: Z. Wang; Jinning: X. Sun; Jinzhou: G. Tao; Jiujiang: Q. Wang; Jun: J. Li; Lankao: X. Guo; Lian Yungang: X. Wang; Liaocheng: K. Zai; Liaoning: Q.
Cui, S. Fan, H. Li, W. Liu, Q. Meng, G. Qi, Y. Qin, G. Wang, N. Wang, G. Xu,
X. Yin, Q. Zhang, S. Zhang, Y. Zhang, Z. Zhang; Liaoyang: R. Liu, F. Wang; Linfen: Y. Zhang; Lingbao: W. K. Li; Lingshou: H.
Zhang; Linyi: X. Xu; Longkou: R.
Ma; Luoyang: F. Guan, T. Yang; Mongolia: R. Zhqo; Nanjing: J. Huang; Nanle: A. Li; Neimeng: J. Zhou; Panjin: X. You; Qi: L. Hao; Qingdao: F. Zhangfang; Qinyuang:
X. Ma; Ruyang: C. Shen; Sanhe:
Q. Li; Shangdong: Z. Hou; Shanghai: N. F. Zho; Shangqiu: G. Huang; Shanxi: P. Guo, J. Lou, Q. P. Wang, Z. Wang; Shenyang: X. Jiang, Z. Li, D. Tian, S. Wang, Z. D. Wu, M. G. Yang; Shi Jiazhuang: Z. C. Li; Taian: S. G. Yang; Tangshan: Y. Tu; Taonan: C. He; Tianjin: Y. Cao, Y. Han; Wangdu:
J. H. Yang; Wangrong: S. Dong; Wuxiang: D. F. Li; Xiang Cheng: Q. F. Zhang; Xianxian: Z. Fan; Xinjiang: D.
Q. An; Xinxiang: J. Liu; Xiping: G. Yang; Xiuwu: X. C. Xu; Xuzhou: Y. Xia; Yantai: H. Xu; Yanzhou: T. Wang; Yichun: D. Li; Yingkou: J. Wei; Yongji: P. Yang; Yuci: C. Y. Liu; Zhengzhou: S. Shang; Zhumadian: Y. G. Zhang; India:Adoni: J. Srinivas, B. Srinivasulu; Ahmedabad: S. Dani, J. Prajapati; Alappuzha:
G. Deepak, J. F. Shallam; Ambur: K. J. Nesaraj; Amritsar: A. Kumar, R. K. Sharma; Annamalainagar: S. Balasubramaniyan, N. Chidambaram, R. Rani; Bangalore: S. Chandra, S. Dwivedi, B. Isaac, R. Kishore, B. J. Kumar,
Y. Kumble, S. Mehrotra, P. R. Nayak, S. S. Ramesh, M. J. Santhosh, P. K. Shetty,
K. Varghese; Bhopal: S. K. Trivedi; Bikaner: R. Beniwal, A. Kalla, R. B. Panwar; Calicut: K. G. Alexander, A. V. Bindu, A. Nambiar; Chennai: D. Barkavi, A. Kalanidhi, T. Pradeep, J. Rajesh, M. Ramesh, S. Shanmugasundaram,
S. Thanikachalam; Cochin: K. K. Haridas, P. Kumar; Doraha: G. Sidhu, R. Singh; Ernakulam: K. N. Pradeep; Ghaziabad: A. Kumar, A. Mittal; Gulbarga: J. B. Bijapure, M. S. Rao; Guntur: N. G. Mohanarjun, M.B. Rao; Hyderabad:
B. R. Babu, N. Dinesh, R. K. Jain, P. A. Jiwani, S. R. Naik, T.N.C. Padmanabhan,
B. S. Raju, R. Rajaram, A. S. V. N. Rao, D. Rao, V. S. P. Rao, B. K. S. Sastry,
S. Sinha; Indore: A. Bharani, G. Verma; Jaipur: R. Gupta, R. K. Tongia, S. Kalra, S. Sharma; Jodhpur: R. Mehrotra, S. Sanghvi, O. P. Soni; Kolkata: A. D. Biswas, A. K. Maity, S. K. Paul; Kottayam: J. Boben, G. Jacob, J. Joseph; Lucknow:
A. Puri, V. K. Puri, H. Singh; Ludhiana: R. Calton,
T. M. Jaison; Meerut: G. K. Aneja; Mumbai: P. G. Kerkar, P. Nyayadhish, P. J. Nathani, S. K. Rane; Nagpur: M. Fulwani, A. S. Jain, P. P. Joshi, A. Khan, U.
K. Mahorkar, R. G. Salkar, A. Somani, R. Wadhwani, S. D. Zawar; Nanded: V. E. Shegokar, S. L. Tungikar; Nashik:
V. Vijan; New Delhi: B. Singh, R. Trehan; Patiala: A. Garg, H. Singh, S. Verma; Pune:
S. Borade, D. Duggal, J. Hiremath; Rohtak: Jagdish, V. K. Katyal, S. B. Siwach; Shimoga: H. R. Devendrappa, J. Narendra, Ratnakar; Thrissur: P. B. Latha, E. B. Manoj, P. P. Mohanan; Trichy: M. Chenniappan, P. Gandhimadhinathan, K. Jeremaiah, B. S. V. Raj, R. Udaysankar; Udaipur: J. K. Chhaparwal, S. K. Kaushik; Vellore: S. T. Chandy, O. George, B. John; Vijayawada: P. Ramesh, V. S. Reddy, P. Srinivas; Vishakapatnam: K. D. Rao, B. R Malipeddi, G. S. R. Murthy; Wardha: R. Joshi, S. P. Kalantri, S. Patil.
Financial Disclosure: Dr Yusuf has received
research grants and honoraria for lectures from Knoll and Abbott Laboratories.
Funding/Support: The study had no external
funding. Reviparin and placebo were donated by Abbott Laboratories.
Role of the Sponsor: Abbott Laboratories did
not participate in the design and conduct of the study; in the collection,
analysis, and interpretation of the data; or in the preparation, review, or
approval of the manuscript.
Additional Information: Slides are available
Acknowledgment: We thank the numerous patients
who volunteered to participate in this trial, Judy Lindeman, for expert secretarial
assistance, and Susan Chrolavicius and Colette Easton, for organizational