Anticoagulant therapy in acute ST elevation myocardial infarction

Authors:
A Michael Lincoff, MD
Donald Cutlip, MD
Section Editors:
Freek Verheugt, MD, FACC, FESC
Christopher P Cannon, MD
Deputy Editor:
Gordon M Saperia, MD, FACC

Contributor Disclosures

All topics are updated as new evidence becomes available and our peer review
process is complete.
Literature review current through: May 2018. | This topic last updated: Mar 15,
2018.

INTRODUCTION — Acute myocardial infarction (MI) results from rupture of an

atherosclerotic plaque, which leads to intraluminal thrombosis. Intraluminal
thrombus impairs distal blood flow and may lead to myocardial ischemia or
infarction. Intraluminal hemostasis is a dynamic process involving both clot
formation and intrinsic fibrinolysis. The goal of antithrombotic therapy (the
combination of anticoagulant and antiplatelet therapy) is to prevent clot
extension and clot reformation in cases where the clot has undergone
fibrinolysis either by intrinsic mechanisms, fibrinolytic treatment, or mechanical
means. (See "The role of the vulnerable plaque in acute coronary
syndromes" and "Overview of hemostasis".)

This topic will review the evidence that parenteral anticoagulant therapy is
beneficial in all patients with acute ST-elevation MI (STEMI) and will provide
recommendations for its use according to whether the patient receives
fibrinolysis, primary percutaneous coronary intervention, or no reperfusion
therapy. Information regarding anticoagulant agents in non-ST elevation acute
coronary syndromes (unstable angina or non-ST elevation MI) and the role of
antiplatelet therapy in STEMI is discussed separately. (See "Anticoagulant
therapy in non-ST elevation acute coronary syndromes" and "Antiplatelet
agents in acute ST elevation myocardial infarction".)

CLASSIFICATION OF ANTICOAGULANT AGENTS — There are three classes

of anticoagulants (figure 1) that have been evaluated in the management of
acute coronary syndromes:

●The heparins, including unfractionated heparin (UFH) and the low

molecular weight heparins (LMWH), are indirect thrombin inhibitors that
complex with antithrombin (AT, formerly known as AT III) and convert AT
from a slow to a rapid inactivator of thrombin, factor Xa, and to a lesser
extent, factors XIIa, XIa, and IXa. (See "Overview of
hemostasis" and"Heparin and LMW heparin: Dosing and adverse effects".)
 There are a number of intrinsic limitations to UFH therapy in patients with
acute myocardial infarction (MI). The most important is that the heparin-
antithrombin complex cannot bind or inactivate thrombin bound within a clot
[1]. Such clot-bound thrombin acts as an important thrombogenic stimulus
at a site of coronary thrombosis, particularly after clot disruption by
fibrinolytic agents [2]. An additional concern with the administration of UFH
has been heparin-induced thrombocytopenia (HIT). The risk is much lower
with LMWH than with UFH and is not seen with the direct thrombin
inhibitors or fondaparinux. (See "Clinical presentation and diagnosis of
heparin-induced thrombocytopenia" and"Management of heparin-induced
thrombocytopenia".)

LMWH inactivates factor Xa, like UFH, but has a lesser effect on thrombin.
As a result, LMWHs do not prolong the aPTT in a predictable fashion. They
have a number of advantages over UFH, including a more predictable
anticoagulant effect and a reduced likelihood of inducing immune-mediated
thrombocytopenia.

●The direct thrombin inhibitors (eg, hirudin, bivalirudin, lepirudin) bind to

and inactivate one or more of the active sites on the thrombin molecule [3].
For patients with ST elevation MI, only bivalirudin is clinically used [4-7].
(See "Direct oral anticoagulants and parenteral direct thrombin inhibitors:
Dosing and adverse effects", section on 'Direct thrombin inhibitors'.)

●The synthetic heparin pentasaccharide fondaparinux acts through

antithrombin to exclusively neutralize factor Xa. (See "Fondaparinux:
Dosing and adverse effects".)

FIBRINOLYTIC THERAPY — We treat all ST-elevation myocardial infarction

(STEMI) patients receiving fibrinolytic therapy with an anticoagulant. In these
patients who will likely receive percutaneous coronary intervention (PCI) after
fibrinolytic therapy, we prefer unfractionated heparin (UFH). For patients who
will not receive PCI, we give either UFH orenoxaparin. Although the evidence
presented below suggests an advantage to enoxaparin in patients treated with
fibrinolytic therapy, we often use UFH since it leaves open the option to proceed
with bailout PCI if there is evidence of failed reperfusion. As discussed below,
we prefer UFH to enoxaparin in STEMI patients undergoing PCI. (See 'Primary
PCI' below.)

We do not recommend fondaparinux for patients treated with fibrinolytic therapy

who may need subsequent PCI. However, it is a reasonable option in patients
who are at high risk of bleeding or those who are unlikely to be referred for PCI.
For these patients treated with fibrinolytic therapy, we suggest anticoagulation
for at least 48 hours and up to eight days.

Bivalirudin has not been evaluated in this setting.

Anticoagulant compared to placebo — We treat all patients receiving fibrinolytic
therapy with anticoagulant, although the evidence to do so in patients receiving
streptokinase is quite weak.

The following information is used to formulate recommendations for the use of

anticoagulant therapy in STEMI patients treated with fibrinolytic therapy:

●There are no randomized trials directly comparing heparin to placebo in

patients treated with aspirin plus a platelet P2Y12 receptor blocker (which is
the recommended antiplatelet therapy). (See "Antiplatelet agents in acute
ST elevation myocardial infarction", section on 'Fibrinolytic therapy'.)

●A 1996 meta-analysis of over 60,000 patients (in six trials) treated with
fibrinolytic therapy (any agent) and aspirin (but not a P2Y12 receptor
blocker) found that the addition of intravenous or high-dose subcutaneous
heparin led to a small and marginally significant reduction in the short-term
risk of death (8.6 versus 9.1 percent) [8]. In addition, the rate of major
bleeding significantly increased in patients treated with heparin.

●Among patients treated with non-fibrin-specific fibrinolytic agents such as

streptokinase or anistreplase, no significant benefit in the outcomes of
infarct vessel patency, mortality, or reinfarction have been found between
those receiving and not receiving UFH [9-11].

●There are conflicting data concerning the efficacy of UFH on infarct vessel
patency in patients treated with alteplase (which is fibrin-specific), with
some studies suggesting a beneficial effect [9,12,13] and others not
[14,15]. Comparison of these studies is difficult due to difference in dosing
and timing of both the fibrinolytic agent and UFH.

There have been no trials comparing low molecular weight heparin (LMWH) to
placebo in patients treated with fibrin-specific agents. However, LMWH has
been evaluated in clinical trials of patients with STEMI treated with
streptokinase. The CREATE trial, the largest of these trials, randomly assigned
15,570 patients with an acute STEMI to reviparin or placebo every 12 hours for
seven days [16]. Fibrinolytic agents (predominantly streptokinase) were
administered to 73 percent of patients. At seven days, the rate of the primary
composite outcome (death, MI, or stroke) was significantly reduced with
reviparin (9.6 versus 11.0 percent; hazard ratio [HR] 0.87, 95% confidence
interval [CI] 0.79-0.95) and this benefit persisted at 30 days. Reviparin
treatment was significantly better when started within two hours after symptom
onset. These findings were confirmed in a meta-analyses of four placebo-
controlled trials, of which CREATE was the largest [11].

UFH compared to LMWH — A number of trials have compared the relative

efficacy of UFH and LMWH (enoxaparin) in patients with STEMI treated with
fibrinolytic agents that were usually fibrin-specific [17-22]. The major initial trials
were ASSENT-3, ASSENT-3 PLUS, and ExTRACT-TIMI 25. In the largest of
these trials, ExTRACT-TIMI 25, the median duration of therapy was two days
for UFH and seven days for enoxaparin.

A 2007 meta-analysis of trials comparing UFH to enoxaparin in over 27,000

STEMI patients receiving fibrinolytic therapy included the above three trials as
well as three smaller studies [23]. The primary end points of death, MI, or major
bleed at 30 days occurred significantly less often in patients receiving
enoxaparin (11.1 versus 12.9 percent, odds ratio [OR] 0.84, 95% CI 0.73-0.97),
with most of the benefit attributable to a reduction in MI. However, major
bleeding occurred significantly more often in patients receiving enoxaparin (2.6
versus 1.8 percent).

Fondaparinux — We do not recommend fondaparinux for patients who receive

fibrinolytic therapy and are likely to undergo early PCI. For these patients who
are not likely to undergo PCI, fondaparinux is a reasonable option to UFH.
However, fondaparinux is not approved for use in patients with STEMI by the
United States Food and Drug Administration.

The efficacy of fondaparinux in STEMI was evaluated in the OASIS-6 trial of

over 12,000 patients who were randomly assigned to fondaparinux
(2.5 mg/day) versus UFH (for those with an indication for heparin) or placebo
(for those with no indication for heparin) for up to eight days [24]. Patients could
be managed with fibrinolytic therapy, primary PCI, or no reperfusion therapy
and they were stratified based on whether they had an indication for heparin.
The results of this complicated trial are reported here; the implications for our
recommendations are discussed in the relevant sections of this topic.

There were two strata:

●Stratum 1 consisted of 5658 patients without planned PCI, in whom

heparin was not indicated. These patients, most of whom received
thrombolytic therapy with streptokinase (78 percent), were randomly
assigned to subcutaneous fondaparinux (initial dose intravenously and then
2.5 mg/day given subcutaneously) or placebo for up to eight days or
hospital discharge, if earlier.

●Stratum 2 consisted of 6434 patients with an indication for heparin (eg,

fibrinolytic therapy with a fibrin-specific agent, primary PCI, or no
reperfusion but eligible for heparin). These patients were randomly
assigned to fondaparinux (initial dose intravenously and then
2.5 mg/day given subcutaneously) for up to eight days or hospital
discharge, or UFH 60 units/kg bolus followed by infusion of
12 units/kg/hr for 24 to 48 hours. Placebos were used to maintain blinding.

Fondaparinux was compared to placebo in stratum 1 and to UFH for 24 to 48

hours in stratum 2. The following findings were noted with fondaparinux at 30
days:
 ●For the entire trial population (Strata 1 and 2), a significant reduction in
the primary end point of death or reinfarction was shown at 30 days (9.7
versus 11.2 percent, HR 0.86, 95% CI 0.77-0.96). This 1.5 percent
absolute benefit was seen by nine days and persisted at 180 days. The
reduction in events was significant for both end points (eg, mortality at 30
days, 7.8 versus 8.9 percent).

●There was a significant reduction in the primary end point of death or

reinfarction in patients in stratum 1 (11.2 versus 14.0 percent with placebo,
HR 0.79) but not in those in stratum 2 who required heparin (8.3 versus 8.7
percent, HR 0.96). The benefit in stratum 1 at 30 days was not seen at two
days. Thus, at least part of the benefit of fondaparinuxin stratum 1 could be
due to the longer duration of antithrombotic therapy (up to eight days for
fondaparinux versus up to 48 hours for UFH) [25].

●On subgroup analysis, patients in stratum 1 who received a non-fibrin

specific agent had a significantly lower rate of the primary end point
with fondaparinux compared to placebo (10.8 versus 13.8 percent); in
patients in stratum 2 who received a fibrin-specific agent, there was no
significant difference between fondaparinux and heparin [26].

●The lack of benefit in stratum 2 reflected a balance between a significant

benefit in patients not undergoing primary PCI (11.5 versus 13.8 percent,
HR 0.82) and a trend toward worse outcomes in patients treated with
primary PCI (6.1 versus 5.1 percent, HR 1.16). Among patients undergoing
primary PCI, fondaparinux was also associated with an increase in guide-
catheter thrombosis and in coronary complications (eg, abrupt closure, no
reflow, dissection). (See 'Fondaparinux' below.)

●In prespecified subgroup analyses, the benefits of fondaparinux were

confined to those receiving fibrinolytic therapy or for individuals not
undergoing reperfusion therapy. No benefit, and a trend toward harm, was
seen in patients undergoing primary PCI [26,27].
(See 'Fondaparinux' below.)

●There was an insignificantly lower rate of bleeding

with fondaparinux compared to heparin (1.7 versus 2.5 percent)

Direct thrombin inhibitors — There are no studies comparing a direct thrombin

inhibitor to placebo. We prefer a heparin to any direct thrombin inhibitor.
(See 'Classification of anticoagulant agents' above.)

Bivalirudin was compared to UFH in the HERO-2 mortality trial of 17,073 STEMI

patients treated with streptokinase [28]. At 30 days, there was no difference
between the two groups in the primary end point of mortality (10.5 versus 10.9
percent with UFH, OR 0.99) and there was a small but significant reduction in
reinfarction at 96 hours with bivalirudin (1.6 versus 2.3 percent). In addition,
there was a nonsignificant trend toward more episodes of severe bleeding (0.7
versus 0.5 percent, p = 0.07) and intracerebral bleeding (0.6 versus 0.4 percent,
p = 0.09) with bivalirudin; there was also a significant increase in moderate and
mild bleeding with bivalirudin.

There are only limited data assessing the role of argatroban or lepirudin [7].
Neither of these drugs is recommended in acute STEMI due to the absence of
large scale trials showing benefit and concerns of an increased rate of bleeding
compared to other anticoagulants. They do have indications for patients with
heparin-induced thrombocytopenia. (See"Management of heparin-induced
thrombocytopenia", section on 'Summary and recommendations'.)

PRIMARY PCI — We recommend anticoagulant therapy for all ST-elevation

myocardial infarction (STEMI) patients undergoing percutaneous coronary
intervention (PCI) based on the evidence from such practice in the broad
population of patients undergoing PCI. Anticoagulation significantly lowers the
risk of acute vessel closure due to thrombosis. (See"Antithrombotic therapy for
elective percutaneous coronary intervention: General use", section on
'Anticoagulants'.)

Anticoagulant therapy should be given as soon as possible once the decision to

proceed with PCI has been made. We prefer unfractionated heparin (UFH) to
other anticoagulants in patients receiving either ticagrelor or prasugrel; in
patients receiving clopidogrel, either heparin or bivalirudin is reasonable.
(See "Antiplatelet agents in acute ST elevation myocardial infarction", section
on 'Primary PCI'.)

This section will summarize the evidence for this recommendation. Dosing and
duration are discussed separately. (See 'Anticoagulant regimens' below.)

UFH compared with bivalirudin — For STEMI patients undergoing PCI with a

radial artery approach who are treated with a potent oral antiplatelet agent
(ticagrelor or prasugrel)and in whom the use of a glycoprotein
[GP] IIb/IIIa inhibitor is not intended, we suggest UFH rather than bivalirudin.
This recommendation in favor of heparin is based on lower cost and the
possibility of a lower rate of definite stent thrombosis. In the following groups of
patients, either UFH or bivalirudin is a reasonable choice:

●Those who receive clopidogrel, rather than prasugrel or ticagrelor;

●Those who are at increased bleeding risk, including women, individuals

with renal dysfunction, and treated using femoral artery catheterization [29].

Each of the trials below, which compared heparin with bivalirudin, differs in the
patients enrolled (some studies included non-ST elevation MI patients), the
P2Y12 receptor blocker used, a radial or a femoral artery approach, the
outcomes measured, and the dose of heparin. These studies are presented in
temporal sequence.
The 2008 HORIZONS AMI was the first major trial comparing bivalirudin with
heparin, and thus it differed from contemporary therapy in that the femoral
access and clopidogrel were used in most patients (as opposed to radial access
and newer generation platelet inhibitors), and heparin was usually given with a
GP IIb/IIIa inhibitor. In HORIZONS AMI trial, 3602 patients were randomly
assigned to either bivalirudin (initial bolus of 0.75 mg/kg followed by an
intravenous infusion of 1.75 mg/kg per hour that was discontinued after PCI)
plus provisional GP IIb/IIIa inhibitor or to UFH (an intravenous bolus of
60 units/kg, with subsequent boluses targeted to an activated clotting time of
200 to 250 seconds) plus planned GPIIb/IIIa inhibitor prior to primary PCI [30].
Approximately 65 percent of patients received UFH prior to randomization. In
the bivalirudin group, 7.5 percent of patients received GP IIb/IIIainhibitor during
PCI for no reflow or giant thrombus. Most patients received a platelet
P2Y12 inhibitor (usually clopidogrel) before catheterization. The following
findings were noted:

●The composite end point of major adverse cardiovascular events (death,

re-infarction, target vessel revascularization, or stroke) occurred at nearly
identical rates by 30 days in the two treatment arms: 5.4 percent
with bivalirudin versus 5.5 percent with heparin and GP IIb/IIIa.

●Major bleeding was significantly reduced by bivalirudin (4.9 versus 8.3

percent, p <0.001). Cardiac death at 30 days occurred significantly less
frequently with bivalirudin (1.8 versus 2.9 percent), as did all-cause
mortality (2.1 versus 3.1 percent).

●The rate of stent thrombosis at 24 hours was higher in

the bivalirudin group (1.3 versus 0.3 percent; p<0.001) but not between 24
hours and 30 days (1.2 versus 1.7 percent).

●The decrease in all-cause mortality with bivalirudin remained significant at

one and three years (3.4 versus 4.8 percent and 5.9 versus 7.7 percent,
respectively) [31]. The survival benefit seen with bivalirudin was not
attributable to its beneficial impact on bleeding [32].

In the 2013 EUROMAX trial, 2218 patients who were being transported for
primary PCI were randomly assigned to the prehospital administration of
either bivalirudin or a heparin (either unfractionated or low molecular weight)
[33]. In the heparin group (n = 1109), nearly 40 percent of patients did not
receive routine GP IIb/IIIa inhibitor (n = 460). The dose of bivalirudin was similar
to that used in HORIZONS AMI, but the drug was continued for four hours after
PCI in an attempt to lower the rate of stent thrombosis seen in HORIZONS AMI.
Other differences with HORIZONS AMI included a significantly greater use of
radial artery access and patients were allowed to receive a more potent
P2Y12 inhibitor (ticagrelor orprasugrel were used in 50 percent of the cases)
than clopidogrel. The following findings were noted:
 ●Bivalirudin reduced the risk of the primary outcome of death or major
bleeding not associated with coronary artery bypass grafting at 30 days
(5.1 versus 8.5 percent; relative risk [RR] 0.60, 95% CI 0.43-0.82), with the
result being driven primarily by a lower risk of major bleeding (2.6 versus
6.0 percent; RR 0.43, 95% CI 0.28-0.66).

●There was no significant difference in the rate of death at 30 days (2.9

versus 3.1 percent, respectively)

●The risk of acute stent thrombosis (within 24 hours) was higher

with bivalirudin (1.1 versus 0.2 percent; RR 6.11, 95% CI 1.37-27.24).

In the 2014 HEAT-PPCI single center trial, 1829 patients undergoing

emergency angiography were randomly assigned to heparin
(70 units/kg) or bivalirudin (bolus 0.75 mg/kg;infusion 1.75 mg/kg/hour) [34].
The rate of GP IIb/IIIa inhibitor use was similar between the two groups (13 and
15 percent, respectively). Ticagrelor or prasugrel was used in nearly 90 percent
of patients. About 80 percent of patients received a radial artery approach for
vascular access. The major difference compared to the HORIZONS AMI and
EUROMAX trials was that a GP IIb/IIIa inhibitor was not routinely used with
heparin.

The following findings were noted at 28 days:

●The primary efficacy outcome (a composite of all-cause mortality,

cerebrovascular accident, reinfarction, or additional unplanned target lesion
revascularization) occurred more often in the bivalirudin group (8.7 versus
5.7 percent; relative risk [RR] 1.52, 95% confidence interval [CI] 0.9-2.13).

●The rate of the primary safety outcome (major bleeding) did not differ
significantly between the bivalirudin and heparin groups (3.5 versus 3.1
percent, respectively; RR 1.15, 95% CI 0.70-1.89).

●Definite or probable stent thrombosis occurred more often

with bivalirudin (3.4 versus 0.9 percent; RR 3.91, 95% CI 1.61-9.52).

The HEAT-PPCI trial called into question the need for the routine use of
GP IIb/IIIa inhibitor in patients receiving heparin plus a potent oral antiplatelet
agent. (See "Antiplatelet agents in acute ST elevation myocardial infarction",
section on 'Glycoprotein IIb/IIIa inhibitors'.)

The open-label 2015 BRIGHT trial randomly assigned 2194 patient with acute
MI (88 percent STEMI) to bivalirudin with a post-PCI infusion (30 minutes to four
hours), heparin alone (100 units/kg), or heparin plus a GP IIb/IIIa inhibitor
(tirofiban) [35]. Study medications were given in the catheterization laboratory.
All patients received 300 to 600 mg of clopidogreland nearly 80 percent of
patients underwent a radial artery approach. A first-generation drug-eluting
stent was used in about 82 percent of cases.
The following findings were noted at 30 days:

●The primary end point of net adverse clinical events, a composite of major
adverse cardiac or cerebral events (all-cause death, reinfarction, ischemia-
driven target vessel revascularization, or stroke), or any bleeding defined
as Bleeding Academic Research Consortium (BARC) types 1 to 5 (table 1)
occurred less often with bivalirudin compared with each of the two heparin
groups (8.8, 13.2, and 17 percent, respectively; RR 0.67, 95% CI 0.50-0.90
and 0.52, 95% CI 0.39-0.69).

●There were no statistically significant differences between treatments in

the 30-day rates of major adverse cardiac or cerebral events (5.0, 5.8, and
4.9 percent, respectively; p = .74),

●The bleeding rates were 4.1, 7.5, and 12.3 percent (p<0.001).

●There was no significant difference in the rates of stent thrombosis.

A contribution from BRIGHT is the observation that continued bivalirudin at PCI

dosing might mitigate the increased risk for acute stent thrombosis that was
seen in the other studies, although this was not confirmed in the subsequent
MATRIX trial. The clinical relevance of the observed reduction of bleeding with
bivalirudin compared with heparin is limited given the high dose of heparin
(100 units/Kg) used in the trial.

The 2015 MATRIX trial randomly assigned 7213 patients with an acute
coronary syndrome (STEMI in 56 percent) to receive either bivalirudin (initial
bolus of 0.75 mg/kg followed by an intravenous infusion of 1.75 mg/kg per hour
that was discontinued after PCI) or unfractionated heparin (70 to 100 units/kg in
patients not receiving GP IIb/IIIa inhibitors and 50 to 70 units per kg in those
receiving them) [36]. In patients receiving bivalirudin, GP IIb/IIIa inhibitors could
only be given to patients with periprocedural ischemic complications; in patients
receiving heparin, GP IIb/IIIa inhibitor (at the discretion of treating physician)
could be given before PCI. Patients in the bivalirudin group were subsequently
randomly assigned to receive or not receive a bivalirudin infusion after the
procedure. All patients received a P2Y12 inhibitor (clopidogrel, ticagrelor,
or prasugrel at the discretion of the treating physician). GP IIb/IIIa inhibitor was
given in 4.6 percent of patients in the bivalirudin group and 25.9 percent of
patients in the heparin group.

The following findings were noted:

●Comparing bivalirudin to heparin, there was no difference in the rates of

the primary end points of major adverse cardiovascular events, a
composite of death, myocardial infarction, or stroke; or net adverse clinical
events, a composite of major bleeding or major adverse cardiovascular
event (10.3 versus 10.9 percent; relative risk 0.94, 95% CI 0.81-1.09 and
11.2 versus 12.4 percent; relative risk 0.89, 95% CI 0.78-1.03,
 respectively). The risk of all-cause death was lower with bivalirudin (1.7
versus 2.3 percent; rate ratio 0.71, 95% CI 0.51-0.99).

●The rate of definite stent thrombosis was borderline higher

with bivalirudin (1.0 versus 0.6 percent; rate ratio 1.71, 95% CI 1.00-2.93).
However, there was no significant difference in the rate of definite or
probable stent thrombosis.

●The rate of BARC 3 or 5 bleeding (table 1) was lower in

the bivalirudin group (1.4 versus 2.5 percent; rate ratio 0.55, 95% CI 0.39-
0.78).The relevance of the finding of reduced bleeding with bivalirudin is
difficult to interpret, given the 25 percent rate of GP IIb/IIIa inhibitor use in
the heparin arm of this study.

Two meta-analyses performed before the VALIDATE-SWEDEHEART

randomized trial, which is discussed below, found that bivalirudin was
associated with a lower rate of major bleeding and a higher rate of stent
thrombosis [37,38].  

In the 2017 open-label VALIDATE-SWEDEHEART randomized trial, 6006

patients with STEMI or non-ST elevation MI undergoing urgent PCI and
receiving treatment with ticagrelor,prasugrel, or cangrelor were randomly
assigned to bivalirudin or heparin [39]. Most patients received radial artery
access, and the use of GP IIb/IIIa inhibitors was not intended. The primary
composite end point (death from any cause, MI, or major bleeding at 180 days)
occurred in 12.3 percent in the bivalirudin group and 12.8 percent in the heparin
group (HR 0.96, 95% CI 0.83-1.10). In addition, no significant difference in the
rate of definite stent thrombosis was found. While not powered to assess the
individual end points in the composite, the trial did not find any apparent
difference between the two anticoagulants for the individual components. The
observed event rates were lower than those predicted.

A 2017 registry study evaluated outcomes in 67,368 STEMI patients (37,708

received heparin and 29,660 received bivalirudin) treated with primary PCI
using radial artery access [40]. GP IIb/IIIa inhibitors were used in 43 and 23
percent of patients, respectively. After adjustment, there was no difference in
the risk for the composite end point of death, MI, or stroke (odds ratio 0.95),
comparing bivalirudin with heparin. However, the rate of stent thrombosis was
higher with bivalirudin (odds ratio 2.11, 95% CI 1.73-2.57).

Contemporary management of STEMI patients who undergo primary PCI

includes the use of potent oral antiplatelet agents and an increase in the use of
radial artery catheterization. Thus, all the studies before VALIDATE-
SWEDHEART do not directly help guide recommendation for contemporary
practice. (See "Antiplatelet agents in acute ST elevation myocardial infarction",
section on 'Our approach'.)
UFH compared to enoxaparin — In patients undergoing primary PCI, we prefer
UFH to enoxaparin. UFH has been compared to enoxaparin in only one small
randomized trial (ATOLL). To the extent that 80 percent of patients received a
GP IIb/IIIa inhibitor, the results cannot be directly applied to cases in which a
GP IIb/IIIa inhibitor is not used.

The open label ATOLL trial randomly assigned 910 STEMI patients
to enoxaparin (intravenous bolus of 0.5 mg/kg) or UFH (initial intravenous bolus
of 70 to 100 units per kg without or 50 to 70 units per kg with
GP IIb/IIIa inhibitor) before primary PCI (67 percent with radial access)
[41]. Aspirin, a P2Y12 receptor blocker (clopidogrel in 93 percent), and
GP IIb/IIIainhibitors (80 percent) were given according to local practice. At 30
days, there was a non-significant reduction in the rate of the primary outcome of
death, complication of MI, procedure failure, or major bleeding with intravenous
enoxaparin (28 versus 34 percent; RR 0.83, 95% CI 0.68-1.01). In addition,
there was no significant difference in the rate of major bleeding between the two
groups.

Important limitations to ATOLL include a high use of GP IIb/IIIa inhibitor

and clopidogrel and its small size. In addition, radial access was used in a high
percent of cases, making the conclusions less certain for patients who undergo
a femoral approach and that it was a relatively small study [42].

Two meta-analyses (one of which included ATOLL) concluded that the use of
low molecular weight heparin (generally enoxaparin) compared to UFH leads to
lower rates of mortality and major bleeding [43,44]. However, important
limitations of these meta-analyses include the absence of patient level data,
variation in the timing, dose, and route of administration of enoxaparin across
studies, and limitations of the individual trials.

For patients in whom bivalirudin will not be chosen, enoxaparin, using the

dosing regimen in the ATOLL trial, is a reasonable alternative to UFH for
patients undergoing PCI with a radial artery approach. When the femoral
approach is used, intravenous enoxaparin may be a reasonable alternative to
UFH, but we are uncertain as to the optimal dose; an intravenous bolus of
0.5 mg/kg is reasonable.

Fondaparinux — As a result of OASIS-6, in which there was a trend toward

worse outcomes with fondaparinux compared to heparin in patients treated with
primary PCI, we recommend not choosing fondaparinux for STEMI patients who
will be treated with primary PCI. (See 'Fondaparinux' above.)

NO REPERFUSION THERAPY — There is some evidence to support the use

of an anticoagulant in ST elevation myocardial infarction (STEMI) patients not
undergoing reperfusion. We believe unfractionated heparin (UFH)
or enoxaparin are reasonable choices; we recommend not
using bivalirudin or fondaparinux in this setting.
Unfractionated heparin — No randomized trials have compared heparin to
placebo among patients who received aspirin but not reperfusion therapy. Two
of the fibrinolytic trials provide conflicting results regarding the efficacy of off-
protocol heparin administration among patients who received aspirin but were
assigned to treatment without a fibrinolytic agent [45,46]. Interpretation of these
conflicting findings is difficult, particularly since the nonrandomized use of
heparin in these trials probably reflects underlying differences in baseline
characteristics and risk factors among the patients treated.

However, systemic anticoagulation should be used if there is evidence of high

risk for systemic or venous thromboembolism (large anterior Q wave MI, severe
left ventricular dysfunction, heart failure, history of systemic or pulmonary
embolus, atrial fibrillation, or echocardiographic evidence of mitral or left
ventricular thrombus).

Low molecular weight heparin — The efficacy of low molecular weight heparin

in this setting was evaluated in the subset of 3325 patients who did not undergo
reperfusion therapy in the CREATE trial discussed in the preceding section [16].
Reviparin was associated with a significant reduction in the incidence of death,
MI, or stroke compared to placebo (15.0 versus 18.3 percent).

The TETAMI trial compared UFH to enoxaparin in 1224 patients who did not
receive reperfusion therapy [47]. The patients were also randomly assigned to
treatment with eithertirofiban or placebo. At 30 days, the incidence of the
combined end point of death, reinfarction, or recurrent angina for patients
treated with enoxaparin was not significantly different from UFH (15.7 versus
17.3 percent).

Bivalirudin — Bivalirudin has not been evaluated in this setting.

Fondaparinux — While there is some evidence to support its use, we do not

use fondaparinux as the anticoagulant in patients treated with no reperfusion
therapy. It is not approved for this purpose by the United States Food and Drug
Administration.

The role of fondaparinux in STEMI patients not treated with reperfusion was
evaluated in a prespecified subgroup analysis of the OASIS-6 trial.
(See 'Fondaparinux' above.) Among the 2867 patients in both strata who were
not reperfused, fondaparinux lowered the rate of the primary efficacy outcome
(death and MI at 30 days) compared to UFH/placebo (12.2 versus 15.1 percent;
hazard ratio 0.80, 95% CI 0.65-0.98) [27]. This subgroup analysis of OASIS-6
provides some support for the use of fondaparinux in patients who are not
reperfused.

ANTICOAGULANT REGIMENS

Dose — The dosing schedules depend upon the clinical situation [48-53]:

●Unfractionated heparin (UFH):

•For patients receiving fibrinolysis, we suggest an intravenous bolus of

60 to 100 units/kg (maximum of 4000 units) followed by
12 units/kg/hour (maximum 1000 units/hour)intravenously to achieve
an activated partial thromboplastin time of 50 to 70 seconds.

•For patients undergoing primary percutaneous coronary intervention

(PCI) and who are treated with heparin, we suggest an intravenous
bolus of 50 to 70 units/kg up to a maximum of 5000 units (target
activated clotting time [ACT] >250 seconds). For patients who receive
a glycoprotein inhibitor, we aim for an ACT of >200 seconds.

•For patients treated with medical therapy (no reperfusion), we suggest

an intravenous [IV] bolus of 50 to 70 units/kg up to a maximum of 5000
units, to be followed by an IV drip of 12 units/kg per hour, with a goal
aPTT of 1.5 to 2 times control or approximately 50 to 75 seconds.

●Enoxaparin:

•For patients treated with fibrinolysis or no reperfusion therapy, and

who are <75 years of age, we use a loading dose of 30 mg IV bolus
followed by 1 mg/kg subcutaneously every 12 hours (maximum of 100
mg for the first two doses only). The first subcutaneous dose should be
administered with the IV bolus. In patients ≥75 years, we use no
loading dose and 0.75 mg/kg subcutaneously every 12 hours
(maximum of 75 mg for the first two doses only). Adjustment for renal
impairment: For patients with a creatinine clearance
<30 mL/minute using the Cockroft-Gault formula and who are <75
years of age, we use a 30 mg IV bolus given with the first dose of the
subcutaneous maintenance regimen, which is adjusted to
1 mg/kg subcutaneously every 24 hours. For patients who are 75
years or older with a creatinine clearance <30 mL/minute, we omit the
intravenous bolus and give a maintenance subcutaneous dose of
1 mg/kg every 24 hours.

Repeated doses of low molecular weight heparin to patients with renal

impairment may lead to accumulation and increased risk of bleeding to
varying degrees. In contrast,unfractionated heparin is not dependent
primarily upon renal function for clearance and is a preferred option for
patients with creatinine clearance <20 mL/min, kidney failure, or
receiving dialysis.

•For patients who will receive PCI after multiple doses of therapeutic
subcutaneous enoxaparin, we give a supplemental IV bolus dose of
0.3 mg/kg of enoxaparin if the last subcutaneous dose was 8 to 12
hours earlier or if only one subcutaneous dose has been administered;
no additional IV bolus dose is given if the last subcutaneous dose was
 within the preceding eight hours [54,55]. 

For patients who will receive PCI more than 12 hours after the last
dose of therapeutic subcutaneous enoxaparin, we prefer UFH [55].

•For patients undergoing primary PCI, we generally prefer UFH. For

patients undergoing primary PCI with a radial artery approach, an
intravenous bolus of 0.5 mg/kg ofenoxaparin (with an additional bolus
of 0.25 mg/kg if the procedure is prolonged by more than two hours)
can be used.

●Fondaparinux:

•For patients being managed medically, we suggest giving 2.5 mg

intravenously, followed by 2.5 mg subcutaneously once
daily. Fondaparinux is not approved by the United States Food and
Drug Administration for this use. Fondaparinux should be avoided in
patients with estimated creatinine clearance less than 30 mL/minute.

•Fondaparinux is not recommended for use in patients undergoing

primary PCI. For patients undergoing PCI who were treated with
fondaparinux, intravenous heparin orbivalirudin should be given during
the procedure to prevent catheter-related thrombosis.
(See 'Classification of anticoagulant agents' above.)

•For patients receiving fibrinolytic therapy, we give 2.5 mg

intravenously as a bolus followed by a subcutaneous dose of 2.5 mg
once daily.

●Bivalirudin – We suggest an initial bolus of 0.75 mg/kg followed by an

intravenous infusion of 1.75 mg/kg per hour. Administration of UFH or low
molecular weight heparin (LMWH) prior to arrival in the catheterization
laboratory for primary PCI is not a contraindication to transitioning to
bivalirudin during the procedure.

Duration — The duration of anticoagulant therapy depends on the initial

management strategy. Although the optimal treatment length has not been
determined for these, the following represent commonly employed regimens in
clinical practice:

●For patients undergoing PCI, heparin is stopped at the end of the

procedure in uncomplicated cases. Some of our experts stop bivalirudin at
the end of the procedure while others continue bivalirudin at a dose of
1.75 mg/kg/hour for four hours after the procedure based on findings from
the EUROMAX trial reported in abstract form [56,57]. In that abstract, the
rate of acute stent thrombosis was lower with this higher dose than with a
dose of 0.25 mg/kg/hour. (See 'UFH compared with bivalirudin' above.)
 ●For patients receiving fibrinolytic therapy or no reperfusion therapy, UFH
is continued for at least two days; if LMWH or fondaparinux are used,
therapy should be continued preferably for up to eight days or until hospital
discharge, whichever occurs earlier.

Continuation of anticoagulation beyond the times suggested above should be

undertaken only if:

●The PCI is complicated and there is an ongoing risk of recurrent ischemia

●There is evidence of high risk for systemic or venous thromboembolism

(eg, anterior STEMI, severe left ventricular dysfunction, heart failure,
history of systemic or pulmonary embolus, or echocardiographic evidence
of left ventricular thrombus) or a pre-existent rational for long-term
anticoagulation, such as patients with prosthetic heart valves or atrial
fibrillation.

RECOMMENDATIONS OF OTHERS — Recommendations for the use of

anticoagulant therapy in patients with ST elevation myocardial infarction
(STEMI) are available in the 2013 American College of
Cardiology Foundation/American Heart Association (ACCF/AHA) [58,59] and
the 2012 European Society of Cardiology (ESC) [60] guidelines for the
management of STEMI, both of which were written before publication of HEAT
PPCI (see 'UFH compared with bivalirudin' above):

●From the ACCF/AHA guideline:

•For patients treated with fibrinolysis, a strong recommendation was

made for anticoagulant therapy in all patients,
with enoxaparin preferred to either unfractionated heparin(UFH)
or fondaparinux. Therapy was recommended for a minimum of 48
hours.

•For patients undergoing primary percutaneous coronary intervention

(PCI), anticoagulant therapy was strongly recommended in all
patients. Bivalirudin was preferred to either UFH or enoxaparin, but the
guideline was published before the results of the HEAT-PPCI trial were
available. (See 'UFH compared with bivalirudin' above.)

•For patients not undergoing reperfusion, no specific recommendation

is made. However, the document acknowledges that fondaparinux is
superior to placebo in these patients.

●From the ESC guideline:

•For patients treated with streptokinase, fondaparinux is preferred; for

patients treated with a fibrin-specific lytic agent, enoxaparin is
preferred to UFH.
 •For patients undergoing primary PCI, bivalirudin is preferred to UFH,
but the guideline was published before the results of the HEAT-PPCI
trial were available. (See 'UFH compared with bivalirudin' above.)

•For patients not undergoing reperfusion, fondaparinux is slightly

preferred to enoxaparin or UFH.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored

guidelines from selected countries and regions around the world are provided
separately. (See "Society guideline links: ST elevation myocardial infarction
(STEMI)".)

SUMMARY AND RECOMMENDATIONS

●All patients with ST-elevation myocardial infarction should be treated with

anticoagulant therapy, which should be given as soon as possible after
diagnosis. The choice of anticoagulant agent depends upon the treatment
strategy for each patient. The dosing and duration for these drugs
(unfractionated heparin [UFH], low molecular weight heparin,bivalirudin,
or fondaparinux) are presented above. (See 'Anticoagulant
regimens' above.)

●For patients treated with primary percutaneous coronary intervention

(PCI), we recommend anticoagulant therapy (Grade 1B). (See 'Primary
PCI' above.)

•We suggest UFH in preference to bivalirudin (Grade 2C). This

recommendation assumes that patients will receive a potent oral
antiplatelet agent (ticagrelor or prasugrel), which we prefer
to clopidogrel. 

For those patients who receive clopidogrel, either heparin or bivalirudin

is a reasonable choice. In addition, patients at high bleeding risk, such
as those treated using a femoral artery approach, are reasonable
candidates for bivalirudin. (See 'Primary PCI' above.) 

●For patients treated with fibrinolytic therapy, we suggest anticoagulant

therapy (Grade 2B). (See 'Fibrinolytic therapy' above.)

•For patients not at high risk of bleeding, we suggest

using enoxaparin as opposed to UFH, fondaparinux,
or bivalirudin (Grade 2C). For those patients in whom PCI is possible
or likely after fibrinolytic therapy, UFH is a reasonable choice.

•For patients at high risk of a bleeding complication and who are not
likely to require PCI, we suggest fondaparinux as opposed
to enoxaparin or UFH (Grade 2C). (See'Fondaparinux' above
and 'Fibrinolytic therapy' above.)
●For patients treated without reperfusion, we suggest anticoagulant
therapy with enoxaparin or UFH as opposed to no anticoagulant therapy,
as soon as possible after presentation (Grade 2B). We do not
use fondaparinux or bivalirudin in this setting. (See 'No reperfusion
therapy' above.)