|
Antithrombotic Therapies in
Acute Coronary Syndrome
Acute coronary syndrome has a high mortality rate, but emerging antithrombin therapies look promising. The authors review the role of thrombin in ACS and explore current therapeutic regimens.
By James M. Schipper, MD, Carl J. Lavie, MD, Arthur G. Grant, MD, Richard V. Milani, MD, and Erik T. Sundell, MD
Acute coronary syndrome (ACS) is a leading cause of emergency department presentations and hospitalizations, representing almost 1.5 million cases annually in the United States. Morbidity and mortality are high for all three types of ACS. Approximately 30% of patients with ST-segment elevation MI (STEMI) die during the first 24 hours of symptom onset and 15% with non-ST-segment MI (NSTEMI) and unstable angina either die or have another infarction within 30 days.
However, recent advances in treatment—particularly antithrombin therapies—promise improved outcomes. Since the 1980s, medical treatment for these conditions has focused on limiting thrombus formation and propagation by inhibiting platelet function and disrupting the coagulation cascade, which stops the progression of the clot or lyses it. This article will focus on the role of thrombin in ACS, the pharmacology of agents that inhibit ACS, and both established and emerging antithrombotic therapies currently used in patients with this syndrome.
UNDERLYING CAUSES OF ACS
To understand the basis for pharmacologic intervention in ACS, it’s essential to understand the pathophysiology of the syndrome. Acute coronary syndrome usually develops when an acute disruption of a vulnerable plaque occurs (see image below). These plaques are characterized by their relatively small size, macrophage infiltration, and thin fibrous cap covering a lipid-rich core. Mechanical factors, as well as inflammation mediated by macrophages, contribute to the weakening and rupture of the fibrous cap. The lipid material found in the plaque and exposed to circulating blood by cap rupture contains high levels of tissue factors, which initiate the coagulation cascade.
 Ruptured atheromatous plaque. This is the most common cause of acute coronary syndrome. |
Less commonly, ACS occurs in the setting of high-grade stenosis, sparked by superficial erosion of a fibrous or calcified plaque. These plaque erosions are less thrombogenic, so the development of ACS may require additional risk factors, such as smoking, hyperglycemia, hypercholesterolemia, or other conditions that increase vascular inflammation.
Regardless of the initiating mechanism, disruption of the plaque sets in motion a complex, self-propagating coagulation cascade, which can quickly form a thrombus capable of partially or completely occluding the coronary artery. Thrombin plays a central role, representing the convergence of the extrinsic and intrinsic pathways and providing positive feedback that amplifies coagulation. Thrombin is initially generated from prothrombin through the enzymatic action of factor Xa and cofactor Va. It further propagates coagulation by activating factors V, VIII, and XI, indirectly promoting its own production. In addition, thrombin converts fibrinogen to fibrin and promotes fibrin’s cross-linking and stabilization by activating factor XIII. Thrombin is also a strong platelet activator, causing the release of adenosine
diphosphate, serotonin, and throm- boxane A2, which further intensify the platelet response and aggregation. Moreover, the glycoprotein (GP) IIb/IIIa receptor is activated by thrombin, aiding in platelet aggregation.
While there are many therapeutic targets, including various platelet receptors, tissue factors, and other elements of the coagulation cascade, thrombin’s central role in fibrin clot formation and platelet activation have made it a primary focus of therapeutic interventions.
ROLE OF ANTIPLATELET THERAPIES
Although this article focuses on antithrombin therapies, it is important to recognize that these therapies are used in conjunction with well-established antiplatelet therapies. For example, almost a decade ago, meta-analyses indicated a nearly 60% reduction in subsequent major ischemic events with the use of aspirin in patients presenting with unstable angina. Long-term studies have demonstrated a 25% major event reduction in post-MI patients treated with aspirin. Therefore, in patients with ACS, we recommend a loading dose of 325 mg of aspirin on presentation (especially for those not currently receiving daily aspirin therapy) and 81 mg daily thereafter. In addition, clopidogrel has been shown to reduce cardiovascular death, nonfatal MI, and stroke by 20% in patients with unstable angina and NSTEMI. It has also caused a 36% reduction in occluded infarct-related arteries in patients with STEMI receiving fibrinolysis and a 10% reduction in mortality following STEMI.
Based on this evidence, most patients with ACS should be started on clopidogrel in the emergency department. The only exceptions are patients whose histories suggest they need immediate bypass surgery. In these cases, clopidogrel therapy should be delayed until decisions about surgery are made. However, most patients with ACS can be treated with percutaneous intervention, intensive medical management, or both.
In addition, clopidogrel therapy has been shown to reduce subsequent major events in patients treated with revascularization, including percutaneous intervention and coronary artery bypass grafting. We recommend using a 600-mg loading dose instead of the 300-mg FDA-approved dose, because this achieves greater antiplatelet effects much earlier without increasing the bleeding risk and has yielded improved outcomes in small trials. A maintenance dose of 75 mg daily should be continued for at least 9 to 12 months following unstable angina, NSTEMI, and STEMI. Although the bleeding risk is increased with the combination of aspirin and clopidogrel, this can be reduced by limiting the aspirin to 81 mg daily.
DRAWBACKS OF HEPARIN
Heparin, also called unfractionated heparin (UFH), is a heterogeneous mixture of highly sulfated cationic glycosaminoglycan polymers with an average molecular weight of 12,000 to 15,000 daltons. Although heparin was first isolated from canine liver cells in 1916, clinical investigation of the drug’s uses did not begin until the 1930s. Heparin inhibits thrombin and factor Xa indirectly through the action of antithrombin, which naturally inhibits the coagulation cascade at multiple levels. Although heparin is a large molecule, a specific pentasaccharide chain located on only 30% to 50% of heparin molecules is what interacts with antithrombin. When heparin binds to antithrombin, this produces a conformational change that enhances the affinity of antithrombin for thrombin and factor Xa, greatly increasing the rate of inactivation. Once the complex is bound to thrombin, it blocks thrombin’s many procoagulant actions, including the conversion of fibrinogen to fibrin.
Unfortunately, heparin has several drawbacks. Due to the large size of the heparin molecule and its binding site, the heparin-antithrombin complex is only able to interact with soluble thrombin. Once thrombin has been incorporated into a clot, the three-dimensional structure of the antithrombin-heparin complex blocks its binding. Therefore, clot-bound thrombin remains active, able to convert fibrin to fibrinogen and activate other clotting factors. Likewise, once factor Xa is platelet-bound, the heparin-antithrombin complex no longer affects it, which supports the need for additional antithrombotic therapies.
A second drawback is heparin’s procoagulant effects. The fact that it is a large molecule predisposes it to having amino acid sequences, which will interact with other proteins, including vitronectin, fibronectin, glycoproteins with high histidine content, and von Willebrand factor. This feature also enables heparin to bind to both thrombin and fibrin simultaneously. In this complex, thrombin continues to function but is shielded from inactivation by antithrombin.
A third drawback is that heparin has affinity for several platelet receptor sites. This can lead to platelet activation and formation of detrimental complexes. The most clinically relevant complex is formed by heparin’s interaction with platelet factor 4, which immediately renders heparin inactive and incapable of binding to antithrombin. However, prolonged exposure to these complexes can stimulate an immune response, which can destroy platelets and lead to the clinical syndrome of heparin-induced thrombocytopenia (HIT). This develops in 1% of patients treated for five days or longer, placing them at greatly increased risk for thrombotic complications.
Still another drawback is that because heparin is dependent on antithrombin, it is ineffective in patients with antithrombin deficiency. Finally, patients on heparin exhibit variable responses and clearance rates because of the drug’s interactions with plasma proteins and complex pharmacodynamics. While weight-based dosing has improved accuracy, patient response is still difficult to predict, so frequent monitoring of partial thromboplastin time is required. This may delay the achievement of therapeutic anticoagulation or cause overanticoagulation.
During the late 1970s and early 1980s, physicians began using heparin in ACS and percutaneous intervention. It was initially given as a single bolus of 3000 units in conjunction with dextran and warfarin. However, as data accumulated, it became clear not only that there was a heterogeneous response to a single-dose heparin bolus, but also that a minimum level of anticoagulation was required to decrease thrombotic events. Not surprisingly, as levels of anticoagulation increased and thrombotic events decreased, bleeding risk also increased. A balance was eventually achieved with the activated clotting time target at 200 to 250 seconds for treatment of ACS and slightly higher for percutaneous intervention. Activated clotting time is used in most cardiac catheterization laboratories instead of partial prothrombin time due to the availability of rapid point-of-care testing.
The first randomized controlled trial comparing aspirin, heparin, and a combination of both to a placebo in the treatment of ACS was published in the late 1980s. Although several small trials and case series followed over the next few years, there are relatively few large trials studying heparin’s efficacy in treating ACS, despite its widespread use. The small trials were examined in a meta-analysis of six randomized controlled trials of 1362 patients, which showed a 33% relative risk reduction (2.9% absolute risk reduction) in death or MI with UFH versus a placebo and a nonsignificant trend toward major bleeding.
ADVANTAGES OF LMWHs
Low-molecular-weight heparins (LMWHs) were developed to help address many of heparin’s drawbacks. Several commercial preparations are available, including enoxaparin, dalteparin, and tinzaparin. Of these, enoxaparin is the most studied and most widely used, with substantial recent data now available. Low-molecular-weight heparins consist of fragmented portions of heparin in various sizes, about a third of which include the pentasaccharide sequence that interacts with antithrombin. Because of their smaller size, they do not form a bridge between antithrombin and thrombin as efficiently as UFH, so their major site of action is at factor Xa.
Low-molecular-weight heparins have several advantages over UFH. In patients with preserved creatinine clearance and normal body mass index, the anticoagulant response is predictable. This obviates the need to monitor the anticoagulation level and allows dosing based solely on weight. A longer duration of action and reliable subcutaneous absorption allow for once- or twice-daily dosing.
Low-molecular-weight heparins also have less interaction with platelet factor 4, greatly decreasing the incidence of HIT.
From a practical standpoint, however, the long half-life of the LMWHs can make gaining arterial access and achieving hemostasis during left-heart catheterization more difficult. This problem is exacerbated by the lack of a rapid anti-Xa activity test in many institutions. Reversal of an LMWH is another potential problem. Although protamine is useful in heparin reversal, its efficacy in reversing an LMWH is variable.
Low-molecular-weight heparins and UFH have been compared as treatments for ACS in several studies. A meta-analysis of six randomized controlled trials and 22,000 patients found no difference in mortality. However, there was a significant reduction in the combined endpoint of death or nonfatal MI with enoxaparin versus UFH (10.1% versus 11.0%, p = 0.02). There has been concern about increased bleeding with enoxaparin; however, this meta-analysis found no difference in major bleeding or number of transfusions.
More recently, the ExTRACT-TIMI 25 trial compared the use of enoxaparin and UFH in patients undergoing fibrinolytic therapy for STEMI. As in previous studies, the combined endpoint of death or MI was less frequent in the enoxaparin group (9.9% versus 12.0%, relative risk 0.83, p = <0.001). In this setting, major and minor bleeding were higher with enoxaparin (2.1% versus 1.4%, p = <0.001, and 2.6% versus 1.8%, p = <0.001, respectively), but there was no increase in intracranial hemorrhage. Overall, compared with heparin, enoxaparin appears to provide increased efficacy with similar or slightly increased bleeding risk, which is offset by substantial decreases in ischemic complications.
Recent analysis of this database has demonstrated that enoxaparin is superior to heparin in subgroups that received percutaneous intervention, as well as those who received clopidogrel. However, some clinicians may choose to use UFH in patients who undergo percutaneous intervention immediately, due to UFH’s rapid onset of action and relative ease of reversal with protamine—especially if vascular injury occurs during the percutaneous intervention procedure.
The ExTRACT-TIMI 25 trial found that in patients with moderate to severe renal dysfunction, enoxaparin and UFH appeared clinically similar, but reducing the enoxaparin dose from 1 mg/kg twice daily to 1 mg/kg once daily is required for patients with a creatinine clearance below 30 ml/minute. Some clinicians prefer UFH in this setting, especially if the treatment is prolonged or an anti-Xa activity test is not readily accessible. The total trial data suggest that the combination of fibrinolytic therapy, aspirin, clopidogrel, and enoxaparin offers an attractive pharmacologic perfusion strategy in patients with STEMI. Reducing the dose of enoxaparin in elderly patients or those with decreased renal function or combining it with fibrinolytics and GP IIb/IIIa inhibitors may further improve its risk/benefit profile, increasing its attractiveness as first-line therapy for patients with ACS. However, many clinicians may decide to use UFH instead of enoxaparin in patients with advanced renal failure.
FONDAPARINUX INHIBITS FACTOR Xa
Fondaparinux is the isolated pentasaccharide sequence that binds to antithrombin. It has no ability to form a link between antithrombin and thrombin and therefore acts exclusively by inhibiting factor Xa. Because of its small, very specific structure, fondaparinux has no risk of causing HIT. Like the LMWHs, it can be administered subcutaneously once daily, without monitoring. It is excreted unchanged in urine, so dosing must be adjusted for renal function. Its long duration of action may make it more difficult to use at institutions that favor an early invasive approach for ACS.
A small dosing study compared UFH with fonda-parinux, 2.5 or 5 mg daily, in patients undergoing percutaneous intervention. Efficacy endpoints showed no difference; however, reduced bleeding occurred with 2.5 mg and a trend toward increased bleeding was seen with 5 mg. For this reason, 2.5 mg daily was used for subsequent trials.
Fondaparinux was compared with enoxaparin in NSTEMI in the OASIS-5 trial. This was designed as a noninferiority trial and met its goal of proving fondaparinux to be noninferior in this setting. The primary endpoint of death, MI, or refractory ischemia at nine days occurred in 5.7% of the enoxaparin group and 5.8% of the fondaparinux group (hazard ratio [HR] 1.01, 95% confidence interval 0.9 to 1.13, p = 0.007 for noninferiority). Additionally, bleeding was reduced with fondaparinux (3.3% versus 7.3%, HR 0.41, p = <0.001), and there was a small but statistically significant reduction in the secondary endpoint of all causes of mortality at 30 days (2.9% versus 3.5%,
p = 0.02) and six months (5.8% versus 6.5%, HR 0.89, p = 0.05).
When fondaparinux and UFH were also compared in the setting of STEMI in the OASIS-6 trial, fondaparinux was found to have a lower incidence of death or MI at 30 days (9.7% versus 11.2%, HR 0.86, p = 0.008). No difference was found in bleeding rates between the two groups. However, in patients who underwent percutaneous intervention, there was a higher incidence of catheter- and intracoronary-related thrombus with fondaparinux, suggesting that fondaparinux alone may not be sufficient, at least not at the 2.5 mg daily dose. Currently, fondaparinux is not indicated for ACS based on the manufacturer’s recommendations and is primarily used for deep vein thrombosis prophylaxis in
hip surgery.
HOW DIRECT ?THROMBIN INHIBITORS WORK
Another solution to the problems posed by heparin is direct inhibition of thrombin. This eliminates dependence on antithrombin, which confers several potential advantages. Direct inhibition of thrombin allows for binding and inhibition of clot-bound fibrin, eliminating an important positive feedback loop in the coagulation cascade. Efficacy is unaffected by antithrombin deficiency. Unlike heparin, direct thrombin inhibitors (DTIs) have no interactions with platelets and also inhibit thrombin-mediated platelet activation.
Direct thrombin inhibitors can be split into two major groups, monovalent and bivalent, based on the number of thrombin binding sites. Bivalent DTIs bind to thrombin’s active site as well as the noncatalytic exosite 1, whereas monovalent DTIs bind only to the
active site.
The use of monovalent DTIs in ACS has been studied much less than the use of bivalent DTIs. A meta-analysis of 11 randomized controlled trials published in 2002, which examined DTIs used in ACS only, included three trials of the monovalent inhibitors argatroban, efegatran, and inogatran. Taken together, these trials showed a reduction in death and MI compared with heparin (4.3% versus 5.1%, p = <0.001), but subgroup analysis showed this effect was due only to bivalent DTIs. Monovalent DTIs (collectively and individually) had no effect on outcome compared with heparin.
Ximelagatran is another monovalent DTI that has received attention for its possible use in atrial fibrillation. It has not yet been studied in ACS but the ESTEEM trial examined its use in post-MI patients. The composite endpoint of death, MI, or severe recurrent ischemia was 13% with ximelagatran versus 16% for placebo (p = 0.032). Currently, production of ximelagatran has been stopped because of concern about drug-induced hepatitis.
The bivalent DTIs include lepirudin and bivalirudin. Lepirudin is the recombinant form of hirudin, which was initially isolated from the salivary glands of leeches. This drug is cleared by the kidneys and dosing must be adjusted for renal function. With normal renal function, its half-life is 60 to 120 minutes. It binds thrombin irreversibly and, unlike with heparin, there is no antidote. Multiple trials have shown that lepirudin is equal in efficacy to heparin for treatment of ACS, but because of the excessive risk of bleeding, lepirudin is unlikely to be approved for this indication.
Currently, bivalirudin is the only DTI approved for use in ACS. This drug is an analog of lepirudin with several clinically important differences. First, enzymatic cleavage of the Pro-Arg bond allows for recovery of thrombin activity. While bivalirudin also has no antidote, its relatively short half-life of about 25 minutes allows for more rapid reversal. Together these properties may explain the decreased bleeding risk compared with lepirudin. Also, bivalirudin is cleared primarily by the liver, so no dosage adjustment is necessary in patients with impaired renal function.
Bivalirudin was first studied in 1995, during the Hirulog Angioplasty Study, where it was compared with UFH in patients undergoing percutaneous intervention for ACS or post-MI angina. This study showed decreased bleeding (3.8% versus 9.8%, p = <0.001) and comparable efficacy results. Recently, bivalirudin has been studied in several ACS trials, with much of the focus on its decreased risk of bleeding, especially compared with combinations of heparin and GP IIb/IIIa. The REPLACE-1, REPLACE-2, and ACUITY trials were performed to address various aspects of this issue.
REPLACE-1 compared bivalirudin and heparin with GP IIb/IIIa receptor inhibition at the physician’sdiscretion in patients undergoing percutaneous intervention for unstable angina or post-MI angina. There was a slight, statistically insignificant trend toward decreased MI, death, or revascularization (5.6% versus 6.9%, p = 0.40) and bleeding (2.1% versus 2.7%, p = 0.52) with bivalirudin. In REPLACE-2, patients were randomized to either bivalirudin with provisional GP IIb/IIIa receptor inhibition or heparin plus planned GP IIb/IIIa inhibition. The primary endpoint of death, MI, urgent repeat revascularization, or in-hospital major bleeding and the secondary endpoint of death, MI, or urgent repeat revascularization were not statistically different. However, in-hospital major bleeding rates were significantly reduced by bivalirudin compared with heparin/GP IIb/IIIa inhibitor-based therapy (2.4% versus 4.1%; p = <0.001).
Economic analysis of the REPLACE-2 trial showed the total cost of the bivalirudin with provisional GP IIb/IIIa inhibitor strategy was slightly less expensive than heparin/GP IIb/IIIa.
The ACUITY study compared UFH or enoxaparin plus a GP IIb/IIIa inhibitor to bivalirudin plus a IIb/IIIa inhibitor to bivalirudin alone. Efficacy and bleeding outcomes were similar for all groups except bivalirudin alone. Bivalirudin monotherapy compared to heparin/GP IIb/IIIa inhibition decreased minor hemorrhage (absolute reduction, 2.1%; p = 0.027) and transfusion (absolute reduction, 4.0%; p = <0.001). There was also a trend toward decreased major hemorrhage in the bivalirudin group compared to the group taking enoxaparin plus GP IIb/IIIa (absolute reduction, 1.5%; p = 0.053).
The ACUITY Timing trial subrandomized the enoxaparin and heparin arms of the ACUITY trial to early use of GP IIb/IIIa inhibition initiated in the emergency department compared to initiation in the cardiac catheterization laboratory. This showed a trend toward decreased ischemic events but also a trend toward increased bleeding. The use of bivalirudin before percutaneous intervention has not been studied.
Because of its hepatic metabolism, bivalirudin may also be safer in patients with renal impairment. A meta-analysis of three trials comparing bivalirudin to heparin in percutaneous intervention also showed that the relative safety benefit of bivalirudin increases with worsening kidney function. Bleeding rates with heparin were inversely proportional to creatinine clearance with heparin, but they remained stable with bivalirudin. Bivalirudin is also indicated to treat hypercoagulability due to HIT.
WARFARIN: SPECIAL CONSIDERATIONS
Patients with ACS requiring therapeutic warfarin due to mechanical valvular prosthesis, recent pulmonary embolism, deep vein thrombosis, atrial fibrillation, and other such conditions will need special considerations. For example, antiplatelet therapy (aspirin and clopidogrel) would still be started for ACS, but if the patient’s international normalized ratio (INR) is therapeutic (greater than 2), heparin or an LMWH would probably not be recommended. In high-risk patients with ACS, aspirin, clopidogrel, and warfarin can be given in combination, but the bleeding risk may be excessive, reaching 5% to 10% per year in some populations, especially the elderly.
Although definitive guidelines for handling these patients is beyond the scope of this article, in some patients we may reduce the dosage of aspirin or warfarin, or both. For example, in patients with atrial fibrillation, we may prescribe 81 mg of aspirin every other day along with warfarin, with the goal of maintaining an INR of 1.5 to 2.5 instead of 2 to 3. However, the decision to make these adjustments requires detailed clinical assessments of thrombotic versus bleeding risk in each patient. The table below lists the mechanisms of action and properties of the various antithrombotic agents used in ACS.
OVERVIEW OF THERAPIES
Heparin reduces death and MI by a modest amount when used in ACS, but it is cumbersome to use and has the serious drawback of potential HIT. The LMWH enoxaparin offers improved outcomes with similar bleeding rates, a decreased risk of HIT, and easier administration. It is currently indicated along with aspirin and clopidogrel for most patients with ACS. Further study is needed before anti-Xa inhibitors are approved for ACS. Bivalirudin is the only DTI approved for use in ACS and is likely to remain so for the near future. It offers decreased bleeding rates with comparable efficacy in higher-risk patients undergoing percutaneous intervention where GP IIb/IIIa inhibition would be indicated with heparin.
Suggested Reading
Antman EM, et al.: Enoxaparin versus unfractionated heparin for ST-elevation myocardial infarction. N Engl J Med 354(14):1477, 2006.
Bittl JA, et al.: Treatment with bivalirudin (Hirulog) as compared with heparin during coronary angioplasty for unstable or postinfarction angina. Hirulog Angioplasty Study Investigators. N Engl J Med 333(12):764, 1995.
Chew DP, et al.: Bivalirudin provides increasing benefit with decreasing renal function: a meta-analysis of randomized trials. Am J Cardiol 92(8):919, 2003.
Cohen DJ, et al.: Economic evaluation of bivalirudin with provisional glycoprotein IIB/IIIA inhibition versus heparin with routine glycoprotein IIB/IIIA inhibition for percutaneous coronary intervention: results from the REPLACE-2 trial. J Am Coll Cardiol 44(9):1792, 2004.
Direct Thrombin Inhibitor Trialists’ Collaborative Group: Direct thrombin inhibitors in acute coronary syndromes: principal results of a meta-analysis based on individual patients’ data. Lancet 359(9303):294, 2002.
Eikelboom JW, et al.: Unfractionated heparin and low-molecular-weight heparin in acute coronary syndrome without ST elevation: a meta-analysis. Lancet 355(9219):1936, 2000.
Fifth Organization to Assess Strategies in Acute Ischemic Syndromes Investigators: Comparison of fondaparinux and enoxaparin in acute coronary syndromes. N Engl J Med 354(14):1464, 2006.
Henry TD: Overcoming heparin limitations in high-risk percutaneous coronary intervention: the alternative strategy-replacing heparin with bivalirudin. J Invasive Cardiol 14 Suppl B:19B, 2002.
Lincoff AM, et al.: Bivalirudin and provisional glycoprotein IIb/IIIa blockade compared with heparin and planned glycoprotein IIb/IIIa blockade during percutaneous coronary intervention: REPLACE-2 randomized trial. JAMA 289(7):853, 2003.
Lincoff AM, et al.: Comparison of bivalirudin versus heparin during percutaneous coronary intervention (the Randomized Evaluation of PCI Linking Angiomax to Reduced Clinical Events [REPLACE]-1 trial). Am J Cardiol 93(9):1092, 2004.
Petersen JL, et al.: Efficacy and bleeding complications among patients randomized to enoxaparin or unfractionated heparin for antithrombin therapy in non-ST-segment elevation acute coronary syndrome: a systematic overview. JAMA 292(1):89, 2004.
Stone GW, et al.: Bivalirudin for patients with acute coronary syndromes. N Engl J Med 355(21):2203, 2006.
Wallentin L, et al.: Oral ximelagatran for secondary prophylaxis after myocardial infarction: the ESTEEM randomized controlled trial. Lancet 362(9386):789, 2003.
Yusuf S, et al.: Effects of fondaparinux on mortality and reinfarction in patients with acute ST-segment elevation myocardial infarction: the OASIS-6 randomized trial. JAMA 295(13):1519, 2006. |
|
