Articles |

Antithrombotic Agents in Coronary Artery Disease FREE TO VIEW

John A. Cairns, MD, Chair; Pierre Théroux, MD; H. Daniel Lewis, Jr., MD; Michael Ezekowitz, MD, FCCP; Thomas W. Meade, DM
Author and Funding Information

Correspondence to: John A. Cairns, MD, Faculty of Medicine, University of British Columbia, 317-2194 Health Sciences Mall, Vancouver, BC, Canada V6T 1Z3.

Chest. 2001;119(1_suppl):228S-252S. doi:10.1378/chest.119.1_suppl.228S
Text Size: A A A
Published online

This article discusses the clinical syndromes of coronary ischemia under the major headings of acute myocardial infarction (AMI), unstable angina, primary prevention, and chronic coronary artery disease. In each section, the greatest weight is given to randomized controlled trials and to overviews of them. Although the overviews give a broad summary picture, they do not provide the immediacy and more direct relevance of individual, large, well-designed clinical trials with unequivocal results. Details of the major trials provide the reader with a sense of the sorts of patients under study and the potential generalizability of the results to the reader’s own patients. This article gives greater emphasis to relatively recent evidence, omitting details of individual trials provided in earlier editions of this consensus conference. The summaries and recommendations are based on all available evidence, whether or not it is discussed in detail in this article.

Systemic Embolization

Systemic embolization after AMI is usually manifested by a stroke. In most cases, the source of the thrombus is the left ventricle (LV), the left atrium, or both.1A number of clinical studies have focused on the incidence of systemic embolism in patients who have experienced AMIs, including the particular risk and the association with echocardiographically detected mural thrombosis in patients experiencing anterior myocardial infarctions (MIs).25 The larger studies indicate a risk of stroke between 1% and 3% for all infarctions, and between 2% and 6% for patients experiencing anterior MIs.

In the prethrombolysis era, echocardiographic studies showed that LV thrombus rarely occurs with inferior AMI2,6 but develops in up to 40% of patients with anterior MIs,4,68 particularly when there is apical akinesis or dyskinesis.57,9 Additional risk factors include extensive infarction, a dilated and dysfunctional LV, acute aneurysm, congestive heart failure (CHF), and atrial fibrillation. Mural thrombi tend to form early after AMI, even during the first 24 h, but may not appear until 1 to 2 weeks after the occurrence of an AMI.7,9A mural thrombus that is protruding into the LV and/or is freely mobile is more likely to embolize.1012 Most systemic emboli occur within the first few weeks following acute infarction,2,6 although a 13% incidence of systemic embolism throughout a 4-year period among patients with associated poor LV function has been reported.13 A few small nonrandomized studies have documented a reduction by anticoagulants of systemic embolisms among AMI patients found to have mural thromboses.2,5,12,1415 The incidence of stroke was reduced in all three of the large short-term trials of anticoagulation in patients who have experienced AMIs.1618 Turpie et al19 randomized 221 patients with anterior MIs to treatment with high-dose heparin (12,500 U subcutaneously [SC] every 12 h) vs treatment with low-dose heparin (5,000 U SC every 12 h). Two-dimensional (2D) echocardiography on day 10 demonstrated a reduction in the rate of mural thrombosis from 32 to 11% (p = 0.0004). No patient had received thrombolytic therapy. An overview of four trials among post-AMI patients15 revealed an odds ratio (OR) of 0.32 (95% confidence interval [CI], 0.2 to 0.52) in favor of anticoagulant therapy for the reduction of mural thrombosis.

In the postthrombolytic era, Kontny et al20 evaluated the efficacy and safety of dalteparin in the prevention of arterial thromboembolism after AMI. A total of 776 patients were enrolled in a multicenter, randomized, double-blind, placebo-controlled trial. Thrombolytic therapy and aspirin were administered to > 90% of the patients. In the placebo group, 21.9% of patients had thrombi compared with 14.2% in the dalteparin group (p = 0.03). The risk of thrombus formation associated with dalteparin was 0.63 (95% CI, 0.43 to 0.92; p = 0.02). There was no significant difference in the rates of arterial embolism, reinfarction, or mortality. Dalteparin was associated with an increased risk of hemorrhage (2.9% vs 0.3%, respectively; p = 0.006). Minor hemorrhage was also more common in the dalteparin group (14.8% vs 1.8%, respectively; p < 0.001). Thus, dalteparin significantly reduced the formation of LV thrombi but at the price of an increased hemorrhagic risk.

The Studio sulla Calciparina nell’Angina e nella Thrombosi Ventriculare Nell’Infarto (SCATI) group21evaluated the use of high-dose SC heparin among AMI patients who received IV streptokinase (SK) but no aspirin. Among patients who had experienced first anterior MIs, the incidence of echocardiographic mural thrombus at hospital discharge was decreased from 36.5 to 17.7% (p < 0.01). A substudy from GISSI-222 showed only a trend to less frequent mural thrombus with a similar SC heparin regimen to that of the SCATI trial, but with no initial IV bolus and with treatment starting at 12 h (22% vs 30% at mean 12 days). There was a trend toward less frequent mural thrombus with SK vs recombinant tissue-type plasminogen activator (rtPA).

Studies undertaken since the use of thrombolytic therapy in AMI became commonplace highlight the link between systemic embolization and atrial fibrillation after MI and its association with poor LV function.2324 In the setting of AMI, atrial fibrillation independently predicts the occurrence of stroke (1.8% vs 0.5%; p = 0.0001) and is associated with three-vessel disease, flow less than Thrombolysis In Myocardial Infarction (TIMI) grade 3, advanced age, larger MI, and worse Killip class. In addition, mortality rates are significantly higher at 30 days (14.3% vs 6.2%; p = 0.001) and at 1 year (21.5% vs 8.6%; p < 0.0001). Thrombolytic therapy appears to reduce the incidence of atrial fibrillation and by inference, the predisposition to develop a stroke.1

Two retrospective analyses of studies designed primarily to evaluate the effect of angiotensin-converting enzyme inhibition of patients with LV dysfunction provided important information with regard to the relationship between ejection fraction and stroke and the potential benefit of antithrombotic therapy.2527 Loh et al25 collected data on 2,231 patients enrolled in the Survival and Ventricular Enlargement study. The incidence of stroke increased as ejection fraction (EF) declined (EF < 28% stroke rate, 8.9%; EF 29 to 35% stroke rate, 7.8%; EF > 35% stroke rate, 4.1%) (relative risk [RR] for EF < 28% vs EF > 35%, 1.86; 95% CI, 0.15 to 3.04; p = 0.01). The protective effect of anticoagulant therapy was evident in all three subgroups (LV EF < 28% RR, 0.17; LV EF 29 to 35% RR, 0.14; LVEF > 35% RR, 0.23). Aspirin was also beneficial among patients with LV EFs < 28% and 29 to 35%. It appears that one or both of these agents should be considered for patients with LV dysfunction, especially for those with LV EFs < 28%.

A retrospective analysis of 6,797 patients from the Studies of Left Ventricular Dysfunction trial27 of patients with EFs < 35% assessed the beneficial roles of warfarin and aspirin. Warfarin was associated with a reduction of all-cause mortality (adjusted hazard ratio, 0.76; 95% CI, 0.65 to 0.89; p = 0.0006). The incidence of fatal stroke was not influenced by warfarin therapy. Antiplatelet drug use was associated with a significant reduction in all-cause mortality (adjusted hazard ratio, 0.8; 95% CI, 0.73 to 0.92; p = 0.0005). The incidence of fatal stroke was not influenced by antiplatelet therapy. Several small studies2,47,10,14 consistently showed an increased risk of systemic embolus with the detection of mural thrombus on 2D echocardiography, and an overview15 demonstrates a pooled OR of 5.45 (95% CI, 3.02 to 7.83). However, because mural thrombus occurs within 48 h,27 it may be inappropriate to rely on detection of 2D echocardiographic evidence of mural thrombosis to prompt the commencement of anticoagulant therapy. It is generally concluded that if heparin treatment is to reduce the embolic risk, it should be initiated as soon as possible following the diagnosis of probable AMI, and it should be maintained until warfarin therapy prolongs the international normalized ratio (INR) to 2 to 3. Evidence of anterior MI should generally be sufficient to prompt the use of heparin. The indication is strengthened in the presence of atrial fibrillation, CHF, dilated LV, acute ventricular aneurysm, or mural thrombus detected on 2D echocardiography.

Thrombus is commonly associated with a chronic LV aneurysm (association, 48 to 66% in surgical studies).2829 However, systemic emboli are infrequent (4 to 5% by preoperative history). In a retrospective study of 89 patients with LV aneurysms, 20 were treated with anticoagulants for 40 patient-years and 69 were not so treated for 288 patient-years.30 Only one patient who was not receiving anticoagulants had a clinical embolic event, an incidence of 0.35 per 100 patient-years. On the basis of this study, the presence of a long-term LV aneurysm, even containing thrombus, does not justify anticoagulant therapy.

Venous Thromboembolism

There is a high incidence of deep venous thrombosis accompanying AMI, which rises with the duration of bed rest, increasing age, and the presence of CHF. The incidence of pulmonary embolism was reduced by almost 50% in the trials of heparin in the preaspirin/fibrinolysis era.3133 In the trials comparing heparin vs no heparin among patients receiving aspirin and fibrinolytic therapy, the baseline incidence of pulmonary embolism is sharply lower than in earlier trials. It is likely that many changes in management contributed to the decrease, in addition to the beneficial effects of aspirin and fibrinolytic therapy. There is only a small incremental benefit of heparin when aspirin and fibrinolytic therapy have been administered.

Although the clinical diagnosis of pulmonary embolism is unreliable as carried out in most trials among patients with AMIs, the findings are supplemented by autopsy data3435 and by the results of studies using full-dose heparin therapy followed by oral anticoagulants, demonstrating a significant reduction in the incidence of venous thrombosis diagnosed by 125I fibrinogen leg scanning.37 It is likely that there is underascertainment of pulmonary embolism, suggesting that the absolute reduction with heparin therapy may be greater than is reported in the trials.

Several well-designed studies have assessed the efficacy of low-dose heparin therapy following AMI. The incidence of venous thrombosis identified prospectively by radiolabeled fibrinogen scanning or venography was reduced by 69% (p < 0.0001).3233 Of 668 patients assessed in these trials for the occurrence of pulmonary embolism, there was a reduction in the incidence from 2.4 to 0.9% (comparison was not significant [NS]). There is no clear difference in the reduction of pulmonary embolism with a low-dose heparin regimen vs a high-dose heparin regimen, but there are very few events to be analyzed in the low-dose trials. To our knowledge, there have been no trials of low-dose heparin vs no-heparin among patients receiving aspirin and fibrinolytic therapy.

Short-term Anticoagulant Trials:

Since 1948, there have been > 30 reports of the use of anticoagulants in patients who have experienced AMIs.3133 However, only three of these trials1618 were of sufficient size to detect a modest but clinically important reduction of mortality. One of these studies found a statistically significant reduction of mortality, while the other two found statistically significant reductions in the incidences of stroke and pulmonary embolism. Design features and outcomes of these trials have been detailed in a prior consensus report.38

Overviews of the anticoagulation trials offer greater perspective and reliability for overall interpretation of the evidence from these trials. Chalmers et al31 reviewed studies of anticoagulants in the management of AMI and identified only six (total patients, 3,854) that allocated treatments randomly. A pooled analysis yielded a reduction of the unweighted mean case fatality rate from 19.6 to 15.4% (risk reduction, 21%; p < 0.05). The thromboembolism rate fell from 21.3 to 11.1% (risk reduction, 48%). Hemorrhage was more frequent among those patients receiving anticoagulants (2.75% vs 8.4%), but there were no hemorrhagic deaths in the randomized trials.

Using rigorous overview approaches, Collins et al3233 summarized the data on early deaths, reinfarction, strokes, pulmonary emboli, and clinically important episodes of bleeding from the 26 unconfounded properly randomized trials of anticoagulant therapy administered to patients in the acute phase of suspected MI (Table 1 ). The trials were categorized as those comparing low-dose SC heparin vs no antithrombotic therapy, high-dose SC heparin vs no antithrombotic therapy, high-dose IV heparin vs no antithrombotic therapy, and high-dose heparin therapy followed by oral anticoagulant therapy vs no antithrombotic therapy. Using these categories, they separately evaluated those trials in which the comparisons were made among patients not receiving aspirin and those among patients receiving aspirin (93% of whom also received fibrinolytic therapy).

It is clear from this overview that in the absence of aspirin and fibrinolytic therapy, heparin reduces deaths by a statistically significant and clinically important amount (ie, approximately 35 fewer deaths per 1,000 patients; p = 0.002). The low-dose SC regimens are probably not effective in reducing death, but there is no clear difference in efficacy among the high-dose regimens. The beneficial effect on mortality is supplemented by reductions of reinfarction (15 per 1,000 patients; p = 0.08), stroke (10 per 1,000 patients; p = 0.01), and pulmonary embolus (19 per 1,000 patients; p = 0.001). The benefits must be balanced against an increase in the incidence of noncerebral major bleeding (13 per 1,000 patients; p = 0.01), which appeared to be confined to the high-dose heparin regimens.

Following successful thrombolysis, there is a risk of infarct-related artery reocclusion of 5 to 30% and a rate of reinfarction of about 4% when acetylsalicylic acid (ASA) is not used.39The thrombolytic agents may paradoxically lead to platelet activation and to the generation of increased amounts of thrombin.40 Hence, there is a good theoretic rationale for the conjoint use of heparin. In the present era, most patients suspected of experiencing an AMI will receive aspirin, and a substantial portion will also receive fibrinolytic therapy. Accordingly, the most relevant trials are those comparing heparin vs no-heparin among patients receiving aspirin and fibrinolytic therapy. In the overview by Collins et al,3233 within the trials of patients receiving aspirin/fibrinolytic therapy, the baseline rates of death, reinfarction, stroke, and pulmonary embolus were markedly lower than those in the preaspirin/fibrinolysis era. Although the addition of heparin led to a reduction of death (5 per 1,000 patients; p = 0.03), reinfarction (3 per 1,000 patients; p = 0.04), and pulmonary embolus (1 per 1,000 patients; p = 0.01), the benefits were small and the statistical significance was marginal. The small mortality benefits observed at 7 days in the International Studies of Infarct Survival (ISIS)-341and GISSI-24243 trials (which contributed most of the patients) became fewer and were no longer statistically significant at 35 days and at 6 months of follow-up. Interestingly, the baseline rate of major bleeding was less than in the preaspirin/fibrinolysis era, and the excess with heparin therapy was less (3 per 1,000 patients; p = 0.0001). Hence, the physician is faced with a modest early benefit of heparin therapy of about five fewer deaths, three fewer reinfarctions, and one less pulmonary embolus, balanced against three more episodes of major bleeding. This suggests that heparin therapy is not indicated for routine use among patients receiving aspirin and fibrinolytic therapy and that a careful selection of patients is warranted (see also the article “Coronary Thrombolysis” in this supplement [see page 228]).

Long-term Anticoagulant Trials:

A series of randomized trials conducted prior to 1980 suggested that long-term anticoagulation therapy might decrease the number of reinfarctions, embolisms, and deaths.4448

Subsequently, the Sixty Plus Reinfarction Study4950 enrolled patients > 60 years of age who had been receiving oral anticoagulation therapy following transmural MIs that had occurred at least 6 months earlier (mean, 6 years) (Table 2 ). The patients were randomly allocated in a double-blind manner to continue treatment with oral anticoagulation therapy (INR, 2.7 to 4.5) or matching placebo. Major and minor extracranial hemorrhages were considerably more frequent in the anticoagulant-treated patients, but transfusion was rare, and there were no fatal hemorrhages. An efficacy analysis of patients adhering to study therapy showed more marked benefits.

The Warfarin Reinfarction Study (WARIS)51 enrolled patients who had sustained AMIs a mean length of time of 27 days previously. They were randomized double-blind to treatment with warfarin (INR, 2.8 to 4.8) or placebo and were advised not to take ASA. There were statistically significant reductions in all-cause mortality, reinfarction, and stroke, and an efficacy analysis revealed more marked benefits. Venous thromboembolism was rare with placebo and did not occur with warfarin. There was one peripheral arterial embolus in each group. There were five intracranial hemorrhages with warfarin treatment, three of them fatal, and there were eight episodes of major extracranial hemorrhage with warfarin treatment, for a combined incidence of major bleeding of 0.6% per year.

The Anticoagulation in the Secondary Prevention of Events in Coronary Thrombosis (ASPECT) research group52enrolled patients who had sustained an AMI within 6 weeks of hospital discharge. They were randomized in a double-blind manner to treatment with acenocoumarol (nicoumalone), phenprocoumon (INR, 2.8 to 4.8), or placebo. There was a favorable trend for the reduction of all-cause mortality with statistically significant reductions in reinfarction and stroke. The combined annual incidence of major bleeding was 1.4% per year with oral anticoagulation and 0.4% per year with placebo. Efficacy analyses showed greater risk reductions with anticoagulation. An overview of these trials reinforces the observations of benefit.53

Neri Serneri et al54 evaluated heparin (12,500 U SC once daily) among 6- to 18-month survivors of Q-wave MI. There was a significant reduction in the rate of reinfarction with favorable trends for the reduction of all-cause and cardiovascular mortality. Efficacy analysis provided stronger evidence for a benefit of heparin. There were no major hemorrhages and no evidence of osteoporosis on bone density measurements.

Antiplatelet Therapy:

Aspirin causes irreversible inhibition of platelet cyclooxygenase, thereby preventing the formation of thromboxane A2, a platelet aggregant and potent vasoconstrictor.55It has no effect on platelet aggregation induced by other agonists and is therefore a weak platelet inhibitor. Although aspirin possesses many other physiologic effects, none of them has been conclusively associated with the benefit observed in thromboembolic diseases. Aspirin inhibits the constitutive cyclooxygenase 1 that is present in almost all cells involved in general homeostasis. It has no effect on cyclooxygenase 2 induced in endothelial cells and in smooth muscle cells following cytokine and growth factor stimulation. The coronary-artery model of Folts et al56 in dogs shows that, before an occlusive fibrin thrombus forms, platelet deposition is reversible with aspirin. The adverse effects of aspirin are primarily related to bleeding, particularly GI. The latter is less common at the low dosage of 80 to 160 mg/d needed to inhibit platelet aggregation. Other nonsteroidal anti-inflammatory drugs reversibly inhibit platelet cyclooxygenase and platelet aggregation, but few clinical trials have been conducted among patients with coronary artery disease.

Various drugs inhibiting thromboxane A2 synthase or blocking the thromboxane A2 receptor, or both, have been investigated in clinical trials. Although they do not decrease prostacyclin production, they have shown no advantage over aspirin.

Sulfinpyrazone, a nonsteroidal anti-inflammatory drug, reversibly inhibits cyclooxygenase, but its exact mechanism of action as a platelet inhibitor remains unclear.55 It also normalizes shortened platelet survival in patients with prosthetic heart valves, and its benefits appear to be more marked on prosthetic rather than natural surfaces. Adverse effects of sulfinpyrazone include exacerbation of peptic ulcer disease, potentiation of the effects of warfarin, and elevation of uric acid.

The effects of dipyridamole appear to be related to an increase in platelet cyclic adenosine monophosphate.55 The antithrombotic effects of dipyridamole are more evident on prosthetic surfaces. Dipyridamole increases shortened platelet survival. In contrast to aspirin, it does not increase the risk of GI bleeding even when combined with warfarin.

Ticlopidine and clopidogrel are adenosine diphosphate (ADP) receptor antagonists that inhibit ADP-induced platelet aggregation and prolong bleeding time.55 The safety profile of ticlopidine is unfavorable, with frequent GI side effects, rash, neutropenia (rarely fatal), thrombocytopenia, and liver function abnormalities (rare).

Clopidogrel has a much more favorable safety profile and is well-tolerated as demonstrated in the Clopidogrel vs Aspirin in Patients at Risk of Ischemic Events study,57 which was conducted among > 19,000 patients.

The glycoprotein (GP) IIb/IIIa antagonists occupy the platelet surface integrin GPIIb/IIIa receptors to prevent fibrinogen binding to platelets, platelet-to-platelet linkages, and platelet aggregation.58Their effects are dose-related but with a steep dose-response curve, narrowing the margin between efficacy and risk of bleeding. Abciximab, a monoclonal antibody against the receptor developed by Coller,59 was the first antagonist available. Peptide and nonpeptide antagonists containing the Arg-Gly-Asp sequence (or Lys-Gly-Asp sequence) for fibrinogen binding were subsequently synthetized. Orally active agents are now available and are being investigated in large clinical trials.

Abciximab is administered as an IV bolus of 0.25 mg/kg followed by an infusion at a rate 10 μg/min for 12 h. This regimen results in > 80% receptor occupancy and inhibition of platelet aggregation induced by ADP. The plasma half-life is short, but because the antibody avidity for the receptor is very high, approximately 70% of receptors are still occupied 12 h after drug therapy discontinuation, in association with 50% inhibition of platelet aggregation. The drug can be found on platelets as long as 15 days after its discontinuation. It is not specific for GPIIb/IIIa (α2β3) but at therapeutic levels also inhibits the integrin vitronectine receptor (αvβ3) that has a role in smooth muscle cell proliferation. Although abciximab is immunogenic, registry data have shown that it can be safely readministered. The administration of abciximab in coronary angioplasty trials was initially associated with significant risk of bleeding, but increased experience with the drug, the use of lower doses of heparin, and early discontinuation of heparin therapy before sheath removal have significantly reduced the risk of bleeding. Abciximab is currently approved for high-risk coronary balloon angioplasty and for the management of acute coronary syndromes in patients for whom a percutaneous intervention is planned in the following 24 h. The experience with its use in patients undergoing stent implantation is favorable. The drug is now being investigated as adjunctive treatment to thrombolysis in patients who have experienced AMIs and in patients with unstable angina and non-Q-wave MI.

Eptifibatide, a cyclic heptapeptide containing the Lys-Gly-Asp sequence for binding to the receptor, and tirofiban, a nonpeptide tyrosine derivative mimicking the Arg-Gly-Asp sequence for fibrinogen binding, are two highly specific synthetic GPIIb/IIIa antagonists.58 The two drugs are approved for the medical treatment of patients with unstable angina and non-Q-wave MI as well as for percutaneous coronary interventions in these patients. Eptifibatide also is approved for use in elective coronary intervention. The two drugs have short half-lives of 2 to 3 h with rapid reversal of the inhibition of platelet aggregation, and they are not antigenic. The dose of eptifibatide recommended in the acute coronary syndromes is a 180-μg/kg bolus followed by a 2-μg/kg/min infusion and for coronary angioplasty a 135-μg/kg bolus followed by 0.5-μg/kg/min infusion The recommended dose of tirofiban is 0.4 μg/kg/min for 30 min followed by an infusion of 0.1 μg/kg/min. Heparin is administered at therapeutic doses with the two drugs. Lamifiban is a nonpeptide IV receptor inhibitor currently being investigated in clinical trials.

Short-term Antiplatelet Therapy Trials:

Elwood and Williams60reported a randomized trial of a single dose of 300 mg aspirin administered by general practitioners on first contact to 2,350 patients thought to have experienced AMIs. It did not show a significant difference in mortality in the 1,750 patients with confirmed AMIs.61The results of the ISIS-2 pilot study62showed decreased mortality with randomization to aspirin or placebo in 619 patients treated with SK for AMIs. Verheught et al63showed a reduction in left anterior descending coronary artery occlusion in 49 patients randomly assigned to aspirin after SK therapy for anterior MIs. They also showed aspirin to be more effective than heparin in the prevention of revascularization, recurrent MI, and death after thrombolytic therapy for AMI in the Antithrombotics in the Prevention of Reocclusion in Coronary Thrombolysis trial.64

ISIS-265 was a randomized, placebo-controlled, double-blind trial of short-term therapy with IV SK, oral aspirin (160 mg daily for 1 month), both, or neither among 17,187 patients with suspected AMIs. In addition to the 23% risk reduction in the 5-week vascular mortality rate among patients receiving SK, there was a 21% reduction among those receiving aspirin, and a 40% reduction among those receiving a combination of SK and aspirin, which are all highly significant reductions. The early reduction in mortality with aspirin persisted when the patients were observed for a mean of 15 months. Aspirin reduced the risk of nonfatal reinfarction by 49%, and nonfatal stroke by 46%. The increased rate of early nonfatal reinfarction noted when SK therapy was used alone is consistent with marked platelet activation after fibrinolytic therapy and was completely resolved when aspirin was added (3.8% vs 1.3%). Aspirin added to the benefit of SK therapy in all groups examined. In particular, among patients > 70 years of age, the combination markedly reduced the mortality rate from 23.8 to 15.8% (p < 0.001) without increasing the risk of hemorrhage or stroke. Because of poor prognosis in older patients who have experienced AMIs, the absolute number of lives saved with aspirin and thrombolytic therapy increases dramatically with age (ie, 2.5 per 100 patients treated who were < 60 years of age, and 7 to 8 per 100 patients treated who were ≥ 60 years of age.

ISIS-2 showed conclusively that short-term aspirin therapy for AMI decreases mortality and reinfarction, has benefits in addition to those of SK, and prevents the increase in reinfarction that occurs after thrombolytic therapy. Consequently, aspirin therapy for patients who have experienced AMIs is not only desirable, but necessary when thrombolytic therapy is used. These benefits were achieved with an aspirin dose of 160 mg/d. Although associated with an increased rate of minor bleeding from 1.9 to 2.5%, aspirin therapy was not associated with any significantly increased risk of major bleeding, including hemorrhagic stroke. The benefit of aspirin, in contrast to that of SK, was independent of the time of onset of treatment. However, early administration seems prudent.

Four pilot studies have been performed with eptifibatide, lamifiban, or abciximab in patients who have experienced AMIs, testing the hypothesis that GPIIb/IIIa antagonism will improve the success of thrombolysis and possibly decrease the risk of bleeding by using lower doses of lytic agents. In the setting of reperfusion, the GPIIb/IIIa antagonists may further improve flow in small arteries by preventing this blockage by thromboembolic material, thereby improving tissue recovery. Large phase III trials are in progress.

Platelet GPIIb/IIIa antagonists also have been evaluated as adjunctive therapy to primary percutaneous transluminal coronary angioplasty (PTCA) for AMI. The Evaluation of Platelet IIb/IIIa Inhibition of Prevention of Ischemic Complication trial66evaluated chimeric c7E3 Fab randomly assigned as a bolus, and as a bolus and 12-h infusion, vs placebo in 2,099 patients undergoing PTCA, of whom 42 had primary PTCA for AMI and 22 had rescue PTCA after failed thrombolysis. Pooling these two groups with AMI, c7E3 bolus and infusion reduced the primary composite end point of death, reinfarction, repeat intervention, or bypass surgery at 30 days to 4.5% compared with placebo (26.1%; p = 0.06). Major bleeding was increased with c7E3 (24% vs 13%, respectively; p = 0.28). At 6 months, ischemic events had decreased from 47.8% with placebo to 4.5% with c7E3 bolus and infusion (p = 0.002). The Randomized Efficacy Study of Tirofiban for Outcomes and Restenosis trial67evaluated tirofiban as an adjunct to PTCA in 2,139 patients with acute coronary syndrome (ie, non-Q-wave AMI or unstable angina). The primary end point of the study was a composite of death, MI, coronary bypass surgery due to angioplasty failure or recurrent ischemia, repeat target-vessel angioplasty for recurrent ischemia, and insertion of a stent due to actual or threatened abrupt closure of the dilated artery. The primary composite end point at 30 days is 12.2% in the placebo and 10.3% in the tirofiban group, a 16% relative reduction (p = 0.160). However, 2 days after undergoing angioplasty the tirofiban group had a 38% relative reduction in the composite end point (p = 0.005), and at 7 days there was a 27% relative reduction (p = 0.002), largely because of a reduction in nonfatal MI and the need for repeat angioplasty. The patients randomly assigned to tirofiban therapy had a similar incidence of major bleeding (2.3%) compared to those assigned to placebo (2.5%). The ReoPro and Primary PTCA Organization and Randomized Trial study68 evaluated 483 patients who underwent primary PTCA for AMI who were randomly assigned to abciximab or placebo. Abciximab was associated with a reduced incidence of death, reinfarction, or urgent target-vessel revascularization compared to placebo at 7 days (3.3% vs 9.9%, respectively; p = 0.003), at 30 days (5.8% vs 11.2%, respectively; p = 0.03), and at 6 months (11.6% vs 17.8%, respectively; p = 0.05). However, there was not a decrease in the 6-month primary end point, which included elective revascularization. Major bleeding also occurred more frequently in the abciximab group than in the placebo group (16.6% vs 9.5%, respectively; p = 0.02)

Ridogrel, which inhibits thromboxane A2 synthase and blocks thromboxane A2/prostaglandin endoperoxide receptors, was evaluated as conjunctive therapy of thrombolysis with SK in RAPT.69 A total of 907 patients who had experienced AMIs were randomly assigned to ridogrel vs aspirin. Patency of the infarct-related vessel, as determined by angiography at 7 to 14 days, was not significantly different between the two treatments, but a post hoc analysis revealed a lower incidence (13%) of ischemic events (eg, reinfarction, recurrent angina, or ischemic stroke) in the ridogrel group compared with the aspirin group (19%; p < 0.025) with no excess of major bleeding.

Long-term Antiplatelet Therapy Trials:

The Antiplatelet Trialists’ collaboration meta-analysis included 145 randomized trials of prolonged antiplatelet therapy vs control in 70,000 high-risk patients with primarily occlusive vascular disease and 30,000 low-risk subjects from the general population.70 Among the high-risk patients, antiplatelet therapy reduced vascular mortality (odds reduction, 18%; SD, 3%; 2p < 0.00001), nonfatal MI (odds reduction, 35%; SD, 4%; 2p < 0.00001), nonfatal stroke (odds reduction, 31%; SD, 5%; 2p < 0.00001), and vascular events (eg, nonfatal MI, nonfatal stroke, and vascular death) (odds reduction, 27%; SD, 2%; p < 0.00001). The risk reduction for vascular mortality was higher in AMI trials (risk reduction, 22%; SD, 4%; 2p < 0.00001) and unstable angina trials than in the prior MI trials (odds reduction, 15%; SD, 5%; 2p < 0.005) and cerebral vascular trials (odds reduction, 14%; SD, 7%; 2p < 0.05).

Among about 20,000 patients who had experienced AMIs, vascular events occurred in 10.6% receiving antiplatelet therapy vs 14.4% on a control regimen (odds reduction, 29%; SD, 4%; 2p < 0.00001). This represented about 38 vascular events avoided per 1,000 patients treated for 1 month. Among 20,000 patients with a history of MI, events occurred in 13.5% receiving antiplatelet therapy vs 17.1% on the control regimen (odds reduction,25%; SD, %; 2p < 0.00001). This represents a 2-year treatment benefit of 36 per 1,000 patients treated. Among 11,000 patients who had experienced prior strokes or transient ischemic attacks (TIAs), events occurred in 18.4% receiving antiplatelet therapy vs 22.2% on control regimen (odds reduction, 22%; SD, 4%; 2p < 0.00001) and a 3-year treatment benefit of 38 per 1,000 patients. The benefit was about 50 events avoided per 1,000 patients for 6 months of treatment among 4,000 patients with unstable angina, and a 1-year benefit of 20 events avoided per 1,000 patients in 16,000 other high-risk patients (2p < 0.00001).

The most widely used antiplatelet regimen in these trials was aspirin. Similar effectiveness was demonstrated with doses of 500 to 1,500 mg daily, 160 to 325 mg daily, and 75 to 150 mg daily. Other studies have demonstrated fewer GI side effects with lower dosages.71 The time required to maximal initiation of platelet aggregation at lower doses supports the use of an initial dose of 160 to 325 mg for acute ischemic syndromes.72

After aspirin, the most widely studied antiplatelet regimens were aspirin plus dipyridamole, sulfinpyrazone alone, and ticlopidine, all of which showed benefits compared with placebo, but there was no clear evidence of benefits different from that of aspirin.

The Anturane Reinfarction Trial7374 of 1,629 patients who were randomized to treatment with sulfinpyrazone or placebo 25 to 35 days after MI reported a reduction of 32% in the cardiac mortality rate (p = 0.058), which was almost entirely due to a reduction in sudden death in the first 6 months. The Food and Drug Administration found fault with the determination of sudden death, and did not approve the claims for sulfinpyrazone.75The Anturane Reinfarction Italian Study76of 726 patients randomized to treatment with sulfinpyrazone or placebo after AMI for a mean of 19 months showed no difference in mortality but did show a decrease in reinfarction and thromboembolic events. The Persantine Aspirin Reinfarction Study-177showed similar mortality and coronary events in the aspirin plus dipyridamole and the aspirin alone groups, consistently lower than in the placebo group, but not statistically significant. The Persantine Aspirin Reinfarction Study-278 showed a statistically significant reduction in the composite values of coronary death and nonfatal MI in the aspirin-plus-dipyridamole group compared to the placebo group. There was no comparison with aspirin alone, thereby providing no evidence of a benefit of aspirin and dipyridamole over aspirin alone.

The Canadian American Ticlopidine Study79and the Ticlopidine Aspirin Stroke Study80showed decreased mortality, stroke, or MI in patients with stroke or TIA randomized to treatment with ticlopidine vs placebo and/or aspirin. The Swedish Ticlopidine Multicentre Study81 showed a decreased incidence of AMI, stroke, or TIA in an on-treatment analysis of patients with intermittent claudication who had been randomized to treatment with ticlopidine vs placebo. In these trials, patients with cerebrovascular disease or peripheral vascular disease had decreased vascular events, including coronary events, while taking ticlopidine. In the Ticlopidine Aspirin Stroke Study,80 the effectiveness of ticlopidine must be weighted against its greater cost and adverse effects compared with aspirin.

CAPRIE57 was a randomized clinical trial of clopidogrel (75 mg daily) vs aspirin (325 mg daily) for a mean of 1.91 years among 19,185 patients who had experienced recent ischemic strokes, recent MIs, or symptomatic peripheral vascular disease. The composite event rate of ischemic stroke, MI, or vascular death was 5.32% with clopidogrel and 5.83% with aspirin, an RR reduction (RRR) of 8.7% (p = 0.043) in favor of clopidogrel. When outcomes were assessed according to the three clinical subgroups, the following RRRs by clopidogrel were found: stroke (RRR, 7.3%; 95% CI, −5.7 to 18.7; p = 0.26), MI (RRR, −3.7%; 95% CI, −22.1 to 12.0; p = 0.66), and peripheral arterial disease (RRR, 23.8%; 95% CI, 8.9 to 36.2; p = 0.0028). There was slightly more, but minimal incidence of, rash and diarrhea in the clopidogrel-treated patients and slightly more bleeding in the aspirin-treated patients. There was no excess of neutropenia or thrombocytopenia.

Trapidil (triazolopyrimidine), an inhibitor of platelet aggregation and a platelet-derived growth factor antagonist, has been studied in 723 patients within 1 month of AMI (Japanese Antiplatelet Myocardial Infarction Study).82 The patients were randomly assigned to aspirin, 80 mg/d, trapidil, 300 mg/d, or no antiplatelet therapy for a mean follow-up period of 475 days. The results seem inconsistent. Recurrent AMI incidence was reduced with aspirin compared to no antiplatelet therapy (p = 0.0045), but the reduction with trapidil was NS (p = 0.0810). The incidence of cardiovascular events (ie, cardiovascular death, reinfarction, uncontrolled unstable angina, and nonfatal ischemic stroke) was reduced in the trapidil group (p = 0.0039), but not in the aspirin group.

Three large trials have been completed with an oral platelet GPIIb/IIIa receptor antagonist in secondary prevention. The Evaluation of Xemilofiban in Controlling Thrombotic Events trial83 evaluated xemilofiban in 7,232 patients undergoing a percutaneous intervention procedure. Patients were randomized to treatment with xemilofiban, 20 mg administered 30 to 90 min before the procedure, and subsequently to either 10 or 20 mg tid or to placebo for a period of 6 months. The primary end point of a composite of death, MI, and urgent intervention occurred at 30 days in 7.3% of patients receiving the high dose, in 8.1% receiving the low dose, and in 8.1% receiving placebo, and at 6 months in 14.1%, 12.6%, and 13.6% of patients, respectively. The mortality rate (low dose, 1.6%; high dose, 1.1%; placebo, 1.0%) was slightly higher in patients receiving the low dose. The Orbofiban in Patients with Unstable Coronary Syndromes-TIMI 16 trial84randomized 10,288 patients who had experienced acute coronary syndrome manifested by chest pain within the previous 72 h with positive cardiac markers, electrocardiographic changes, or prior cardiovascular disease to orbofiban 50 mg bid for the duration of the trial, orbofiban 50 mg bid for 30 days followed by 30 mg bid, or to placebo. The treatment phase was planned to last an average of 1 year, but recruitment of patients was stopped prematurely because of excess 30-day mortality with orbofiban, 2.0% for the two orbofiban groups combined, vs 1.4% for the placebo group (p=0.02) and 2.3% for the group receiving orbofiban 50 mg followed by 30 mg (p=0.04 vs placebo). The primary outcomes of death, MI, recurrent ischemia leading to hospitalization or urgent intervention, or stroke occurred through 10 months in 23.1% of patients receiving orbofiban 50 mg followed by 30 mg, 22.9% of patients receiving orbofiban 50 mg bid throughout the trial, and 22.8% of patients receiving placebo and the outcome of death in 5.1%, 4.5%, and 3.7% of patients, respectively. The Sibrafiban versus Aspirin to Yield Maximum Protection from Ischemic Heart Events Post-Acute Coronary Syndromes trial85 directly compared the GPIIb/IIIa antagonists to aspirin in 9,233 patients randomized within 7 days of experiencing an acute coronary syndrome to treatment with aspirin (80 mg bid) or to low-dose sibrafiban to achieve at least a 25% inhibition of platelet aggregation, or to high-dose sibrafiban to achieve 50% inhibition. The primary end point of the composite of death, nonfatal infarction or reinfarction, or severe recurrent ischemia at 90 days did not differ significantly among the groups assigned to aspirin (9.8%), to low-dose sibrafiban (10.1%), and to high-dose sibrafiban (10.1%). A second Sibrafiban versus Aspirin to Yield Maximum Protection from Ischemic Heart Events Post-Acute Coronary Syndromes trial testing regimens of low-dose sibrafiban with aspirin and high-dose sibrafiban without aspirin was interrupted following the completion of the first trial. The BRAVO (Blockade of the Receptor to Avoid Vascular Occlusion trial) is currently ongoing, evaluating lefradafiban in a large population of patients who have experienced recent acute coronary syndrome or stroke or who have peripheral vascular disease. So far, the trials with oral GPIIb/IIIa antagonists have not shown a benefit but have shown an excess of bleeding and a trend to increased mortality.

Antiplatelet therapy is effective in reducing the number of vascular events (ie, nonfatal MI, nonfatal stroke, or vascular death) in patients with evidence of atherosclerotic disease (ie, acute MI, unstable angina, history of AMI, stroke, TIA, stable angina, or peripheral vascular disease). Aspirin, 75 to 325 mg/d, is as effective as any other regimen, with little risk of adverse events such as GI or intracerebral hemorrhaging. Its benefits are seen regardless of age or sex. Aspirin therapy in these high-risk patients reduces vascular events by about one fourth, nonfatal MI by one third, nonfatal stroke by one third, and vascular death by one sixth.

Comparisons of Antiplatelet and Anticoagulant Therapy:

Oral anticoagulation has been compared directly with aspirin in several trials (Table 3 ). The German-Austrian trial86enrolled 942 patients within 30 to 42 days of their experiencing AMIs and assigned them to aspirin, placebo, or phenprocoumon therapy. Over a 2-year follow-up period, the aspirin-treated patients had statistically insignificant reductions of 26% for all-cause mortality and 46.3% for coronary mortality in comparison to phenprocoumon. Aspirin showed a favorable trend in comparison to placebo, but phenprocoumon did not. In the Enquête de Prévention Secondaire de l’Infarctus du Myocarde trial,87 1,303 patients were randomized a mean of 11.4 days following AMI to treatment with aspirin or one of several anticoagulants. Over a mean follow-up period of 29 months, the all-cause mortality rate was 10.3% with anticoagulation and 11.1% with aspirin. The study was stopped early when it appeared that a statistically significant lower mortality rate with aspirin would not be found.

The Aspirin/Anticoagulants Following Thrombolysis with Anistreplase (Eminase) and Recurrent Infarction (AFTER) study88 enrolled 1,036 survivors of AMI who had received anistreplase (APSAC). They were randomized to treatment with anticoagulation (IV heparin followed by warfarin or other oral anticoagulant) or aspirin (150 mg daily) and were followed up for the principal outcome of cardiac death or recurrent MI by 30 days. The rates of the principal outcome were 11.0% with anticoagulation and 11.2% with aspirin. The trial was stopped early because of a declining enrollment rate and was underpowered to rule out a difference between the two therapies. However, the rate of severe bleeding or stroke was significantly higher with anticoagulation therapy than with aspirin (3.9% vs 1.7%, respectively; OR, 0.44; 95% CI, 0.20 to 0.97; p = 0.04).

The Coumadin, Aspirin, Reinfarction Study study89compared aspirin, 160 mg, with warfarin, 1 mg, plus aspirin, 80 mg, and with warfarin, 3 mg, plus aspirin, 80 mg, in a randomized double-blind study of 8,803 patients enrolled 3 to 21 days after experiencing MIs. During a median follow-up period of 14 months, the primary outcome composite of reinfarction, nonfatal ischemic stroke, or cardiovascular death occurred with a rate of 8.6% in the 160-mg aspirin group, 8.8% in the 1-mg warfarin plus 80-mg aspirin group, and 8.4% in the 3-mg warfarin plus 80-mg aspirin group. Major hemorrhage occurred with rates of 0.74% in the aspirin group and 1.4% in the 3-mg warfarin/80-mg aspirin group. Among 3,382 patients assigned to 3 mg warfarin/80 mg aspirin, the INRs were 1.51 at week 1, 1.27 at week 4, and 1.19 at 6 months. The authors concluded that low fixed-dose warfarin therapy (1 or 3 mg) combined with low-dose aspirin therapy (80 mg) did not provide clinical benefit beyond that achievable with 160 mg aspirin. The results are consistent with the body of literature suggesting that warfarin is effective only at INR ranges between 2 and 3.5, at least in the short term. The results of the Thrombosis Prevention Trial (TPT)90 suggest that warfarin therapy at a lower INR (approximately 1.5) may be beneficial in primary prevention.

The Combined Hemotherapy and Mortality Prevention Study (CHAMP)91 was an open-label Veterans Administration (VA) cooperative study trial that sought to demonstrate a 15% reduction in all-cause mortality in survivors of AMI treated with combined therapy (ie, warfarin, INR 1.5 to 2.5, plus aspirin, 81 mg) compared to aspirin therapy (162 mg) alone.

The trial was conducted in 78 VA Medical Centers. Patients were recruited within 14 days of experiencing acute MIs. The study population was derived from 20,036 subjects who were screened and consisted of 5,059 subjects who were eventually randomized. The major reasons for exclusion were lack of consent (24%), alternative indication for warfarin (16%), serious comorbidity (15%), bleeding risks (12%), and patients evaluated beyond the 14-day time window (5%). The study population consisted mostly of men with a mean age of 62 years. Approximately half had hypertension, had experienced angina previously, and were current smokers, and about 35% had had an anterior MI. Twenty-seven percent were diabetics, 9% had experienced a stroke previously, and 8% had a history of heart failure. Vital status was known in all but 59 patients (aspirin group, 26; combination, 33). There were 6,940 patient-years of follow-up in the aspirin group, and 6,789 patient-years in the combination group with a median follow-up period of 2.75 years in both groups. Therapy was initiated in 91% of patients in the aspirin patients and 81% in the combination therapy group, and was discontinued in 13% of patients in the aspirin group and 26% in the combination therapy group. Using an intention-to-treat analysis, there was no significant difference in the total mortality rate (17.3% vs 17.3%, respectively), cardiovascular mortality (4.7% vs 4.2%, respectively), nonfatal stroke (4.7% vs 4.2%, respectively), and nonfatal MI (13.1% vs 13.3%, respectively). Major bleeding, mostly GI, was more common in the combination therapy group than in the aspirin group (combination therapy group, 1.25 major episodes of bleeding per 100 patient-years; aspirin-alone group, 0.69 major episodes of bleeding per 100 patient-years). Intracranial hemorrhages occurred at a 0.2% rate for both groups, and episodes of fatal bleeding were no different between the two groups. It was concluded that there was no survival advantage for adding warfarin to aspirin therapy in survivors of AMI. Combination therapy did not reduce rates of nonfatal MIs, nonfatal stroke, or vascular mortality. Major bleeding complications in both groups were low and were primarily GI. There was a nearly twofold increase in major hemorrhage in the combination therapy group. The mean INR in this study was 1.9.

Anand and Yusuf92 recently published a meta-analysis of anticoagulation therapy in patients with coronary artery disease. Since it has been recognized for a long time that the therapeutic window for oral anticoagulation is narrow, investigators divided their analysis of anticoagulation control into those patients who had received high-intensity anticoagulation therapy (INR, between 2.8 and 4.8), moderate-intensity anticoagulation therapy (INR, 2 to 3), and low-intensity anticoagulation therapy (INR, < 2.0). In comparisons of anticoagulation plus aspirin vs aspirin alone, they were classified as moderate-to-high-intensity anticoagulation therapy (INR, ≥ 2) and low-intensity anticoagulation therapy (INR, < 2.0). In the CHAMP study, the INR range was 1.5 to 2.5 (mean, 1.9). While the analysis of anticoagulation therapy by Anand and Yusuf92 involved patients with coronary disease and was not confined to those who had experienced AMIs, the data set was similar to the patients enrolled into the CHAMP trial. The majority of patients began therapy within 3 months of a hospitalization, presumably for an MI. The major finding of this meta-analysis was that moderate-intensity and high-intensity anticoagulation therapy were effective in reducing the incidence of MI and stroke compared to control subjects but carried a severalfold increased risk of bleeding. Moderate-to-high-intensity anticoagulation therapy plus aspirin compared to heparin therapy alone in three trials involving a small population of patients (n = 480) reduced the incidence of the composite of death, MI, or stroke by 50%. The benefit remains uncertain with so few patients. As in both the CHAMP and the Coumadin, Aspirin, Reinfarction study, low-intensity anticoagulation therapy plus heparin was not superior to aspirin therapy alone.


The 2D echocardiographic studies indicate that the risk of systemic embolism is related to the presence of mural thrombus that occurs in up to 40% of patients who have experienced transmural anterior MIs but is very uncommon with patients who have experienced inferior MIs. Patients with extensive LV dysfunction and CHF, a history of previous embolism, and atrial fibrillation are at particularly high risk. A definitive overview of randomized trials showed that in the pre-ASA/fibrinolytic era, heparin (high-dose SC or IV) reduced mortality by 35 deaths per 1,000 patients, while reducing the incidence of reinfarction, stroke, and pulmonary embolism and increasing the incidence of noncerebral major bleeding. The review also showed that among patients receiving ASA/fibrinolytic therapy, there was a much more modest mortality reduction of 5 deaths per 1,000 patients, and only minor reductions of reinfarction and pulmonary embolus, but with very few episodes of excess major bleeding. Accordingly, among those patients who had experienced transmural anterior MIs, extensive LV dysfunction and CHF, a previous embolism, or atrial fibrillation who did not receive fibrinolytic therapy, there is a strong case for early full-dose heparin therapy (ie, a bolus of 60 U/kg IV to a maximum dose of 4,000 U, an initial maintenance dose of 12 U/kg/h to a maximum of 1,000 U, activated partial thromboplastin time [APTT] 1.5 to 2 times that of the control group). IV heparin therapy should be sustained throughout the hospital stay or replaced by SC heparin (initial dose, 17,500 U every 12 h; APTT, 1.5 to 2 times that of the control group) or oral warfarin (INR, 2.0 to 3.0). A low-dose SC heparin regimen (7,500 U bid) may be reserved for patients with non-Q-wave or inferior infarction who are free of CHF, previous embolism, and atrial fibrillation.

Patients suspected of having Q-wave MIs generally should receive aspirin and fibrinolytic therapy. Although the benefits of heparin are much less than when no fibrinolytic therapy is given, the evidence suggests that patients who are to receive rtPA should also receive heparin as a 75-U/kg IV bolus at the time of initiating the rtPA infusion, with an initial maintenance infusion of 1,000 to 1,200 U/h to maintain APTT at 1.5 to 2 times the control rate. A 48-h infusion is likely to be sufficient if aspirin is being given, and high-dose heparin therapy should be sustained only if there appears to be a high risk of systemic embolism (eg, large anterior MI, CHF, previous systemic embolus, or atrial fibrillation). Otherwise, only low-dose heparin therapy (7,500 U SC every 12 h) is indicated for prophylaxis against venous thrombosis until the patient is fully ambulatory. By extrapolation from the rtPA data, a similar approach to the use of anticoagulation may be recommended for those patients who have received reteplase or Tenecteplase (TNK). If the patient has received SK or APSAC, high-dose heparin should be administered only if there appears to be a high risk of systemic embolism. In such patients, the APTT should be evaluated at intervals beginning about 4 h after the initiation of thrombolytic therapy, and heparin therapy should be initiated when the APTT falls into the therapeutic range.

The appropriate duration of anticoagulant therapy is uncertain. The early large trials of full-dose anticoagulation in patients with AMIs required oral anticoagulation for approximately 1 month following the AMI. The risk of systemic embolism following MI persists beyond the hospital stay but appears to fall off after 2 to 3 months.2 Studies of the subsequent survival of patients discharged from the hospital alive following AMIs indicate that the mortality risk is highest in the first few months following an MI, falling to a much lower rate by 1 year.9394 Three well-designed clinical trials of long-term oral anticoagulation4952 following AMI have shown reductions of death, reinfarction, and stroke with acceptable hemorrhagic risks, although the INR target ranges were higher than is now recommended in most antithrombotic regimens (Sixty Plus reinfarction studies INR, 2.7 to 4.5; WARIS and ASPECT INR, 2.8 to 4.8).

The data argue for a period of oral anticoagulation extending somewhat beyond the hospital stay for those patients with transmural anterior MIs or extensive LV dysfunction and CHF. The data from recent studies of nonrheumatic atrial fibrillation support permanent anticoagulation for patients with atrial fibrillation following MIs.95 Despite the benefits of long-term warfarin therapy for the reduction of vascular events among MI survivors, aspirin will probably continue to be the agent of choice because of its simplicity, low cost, and safety. As yet, and to our knowledge, no clinical trials have demonstrated a benefit for one agent over the other. When warfarin is considered necessary for a patient, specifically for the reduction of the risk of venous or systemic embolism, aspirin need not be given concurrently, and treatment with it should be recommended only when the warfarin therapy is discontinued.

The use of antithrombotic therapy in patients with unstable angina and non-ST-segment elevation MIs has evolved at an accelerated pace in the last decade. Large trials have been completed with low-molecular-weight heparins (LMWHs), direct thrombin inhibitors, and antagonists of the GPIIb/IIIa receptor. The LMWHs have been shown to be as effective or more effective than unfractionated heparin, to provide more reliable anticoagulation, and to add convenience for use. The direct thrombin inhibitors have shown an advantage over unfractionated heparin during the acute phase but less benefit long term, whereas the GPIIb/IIIa antagonists were introduced to clinical practice.

Antiplatelet Agents

Table 4 summarizes the results of all trials comparing an oral antiplatelet agent to placebo in unstable angina. These trials have been discussed in detail in the previous consensus conference publications.96100 Most have used aspirin and have demonstrated a significant risk reduction in the rates of all-cause mortality or cardiac death and nonfatal MI. The results were homogenous among the trials, despite variations in the dose of aspirin, the time of initiation after the acute clinical episode, and the duration of follow-up. The drug reduced the risk of fatal or nonfatal MI by 71% at 7 days,9899 by 60% at 3 months,96and by 50% at 2 years.97 In the Antiplatelet Trialists’ collaboration meta-analysis,70 vascular events after 6 months in 4,000 patients randomized with unstable angina were reduced from 14 to 9% (p < 0.00001). The odds of MI, stroke, or vascular death were reduced by > 25%. Other antiplatelet agents that have been tested also have shown, in general, a similar benefit, except for sulfinpyrazone, which showed no benefit over placebo and no favorable interaction with aspirin in one clinical trial.97 One randomized, double-blind study of triflusal, an analog of aspirin with somewhat different pharmacokinetic properties, showed in 281 patients a 54.4% reduction in the 6-month risk of death or MI compared with placebo (12.3% vs 5.6%; p = 0.048).101Ticlopidine, 250 mg bid, tested against conventional treatment excluding aspirin in an open-label study of 652 patients with unstable angina, reduced the rate of vascular death or nonfatal MI by 46.3% from 13.6 to 7.3% (p = 0.009).102 The life table analyses showed that the gain with ticlopidine appeared after a period of 10 days of treatment, making the treatment less suitable for the acute phase. Clopidogrel has not yet been tested in patients who have experienced unstable angina/non-ST-segment elevation MI. An ongoing trial is investigating the potential benefit of the combination clopidogrel and aspirin vs aspirin alone.

The results of the trials with IV GPIIb/IIIa antagonists in patients with unstable angina are shown in Table 5 . The trials of abciximab have been performed, in the main, among patients undergoing percutaneous intervention but have enrolled a large proportion of patients with an acute coronary syndrome. One of them, the c7E3 Fab Antiplatelet Therapy in Unstable Refractory Ischemia (CAPTURE) trial, exclusively enrolled patients with unstable angina and refractory angina who were receiving medical management after the angiographic identification of a culprit lesion suitable for coronary angioplasty.103104 A total of 1,265 patients were randomized to treatment with placebo or abciximab for 18 to 24 h before the procedure and for 1 h after. By 30 days, the rate of the composite outcome of death, MI, or urgent intervention was reduced with abciximab from 15.9 to 11.3% (p = 0.012). At the 6-month follow-up, however, death, MI, or repeat intervention had occurred in 193 patients in each group. The odds of death or nonfatal MI were still significantly reduced in patients with elevated baseline troponin T levels (0.32; 95% CI, 0.14 to 0.62; p = 0.002).104

Eptifibatide, lamifiban, and tirofiban were evaluated in trials among patients with unstable angina or non-Q-wave MI. The Platelet IIb/IIIa Antagonsim for the Reduction of Acute Coronary Syndrome Events in a Global Organization Network trial105 randomized 2,282 patients to the following five groups: placebo plus heparin; low-dose lamifiban with heparin; low-dose lamifiban without heparin; high-dose lamifiban with heparin; or high-dose lamifiban without heparin. The composite primary outcome of death or MI at 30 days occurred in 11.7% of patients receiving standard therapy, in 10.6% receiving low-dose lamifiban, and in 12% receiving the high dose. By 6 months, the composite outcome was the least frequent among the patients assigned low-dose lamifiban (p = 0.027). There was excess bleeding in the group receiving high-dose lamifiban plus heparin (Table 5 ).

The Platelet Receptor Inhibition for Ischemic Syndrome Management (PRISM) trial106 directly compared tirofiban with heparin in 3,231 patients who had experienced chest pain at rest within the previous 24 h and had ST-T-segment changes, elevation of levels of creatine kinase (CK) or CK-MB, or previously documented coronary artery disease. The drug was administered for 48 h. Interventions were performed when indicated after the discontinuation of tirofiban therapy. The primary composite end point of death, MI, and refractory ischemia measured at the end of the 48-h infusion period was decreased from 5.6% with heparin to 3.8% with tirofiban (RR, 0.67; 95% CI, 0.48 to 0.92; p < 0.01). The reductions at 1 week and 1 month were not statistically significant. However, the rate of death was significantly less with tirofiban (2.3% vs 3.6%, respectively; p = 0.02) at 30 days. Tirofiban had lowered the risk of death at 30 days by 75% (p = 0.004) and lowered the risk of MI by 63% (p = 0.01) in patients with elevated levels of troponin T or troponin I at baseline,104,106108 which is similar to the observation in the CAPTURE trial.104

The PRISM-PLUS trial108and the Platelet IIb/IIIa in Unstable Angina Receptor Suppression Using Integrilin (PURSUIT) trial109 evaluated GPIIb/IIIa antagonists in an overall management strategy that included medical therapies and inventions that were indicated while patients were receiving the study drug. The following entry criteria in these two trials were similar: chest pain at rest (PRISM-PLUS trial, within 12 h; PURSUIT trial, 24 h) plus ST-T-wave changes; or the isoenzyme of CK with muscle and bone subunits elevation. In the PRISM-PLUS trial,108a total of 1,915 patients were randomized to treatment with tirofiban alone, heparin alone, or tirofiban plus heparin for a mean duration of 71 h. In the PURSUIT trial,109 10,948 patients were randomized to treatment with a low or high dose of eptifibatide or placebo for a mean duration of 72 h; it was recommended that aspirin and IV heparin be administered concomitantly. The tirofiban-alone arm in the PRISM-PLUS trial was dropped prematurely because of an excess mortality rate at 7 days; the low-dose eptifibatide arm was discontinued in the PURSUIT study, as predefined in the absence of significant adverse events with the high dose. In the PRISM-PLUS trial, tirofiban plus heparin vs heparin alone reduced the primary composite end point of death, MI, or refractory ischemia at 7 days from 17.9 to 12.9% (RR, 0.68; 95% CI, 0.53 to 0.97; p = 0.004). The reductions were also statistically significant at 30 days (18.5% vs 22.3%; p = 0.03) and at 6 months (27.7% vs 32.1%; p = 0.02). The composite of death or MI was reduced from 8.3 to 4.9% at 7 days (p = 0.006), from 11.9 to 8.7% at 30 days (p = 0.03), and from 15.3 to 12.3% at 6 months (p = 0.06). In the PURSUIT trial, the eptifibatide group, compared with the placebo group, had a 1.5% absolute reduction in the incidence of the primary end point of death or MI at 30 days (eptifibatide group, 14.2%; placebo group, 15.7%; p = 0.04) with benefit maintained at the 6-month follow-up.

These trials have all independently documented an early reduction with the GPIIB/IIIa antagonists in the rate of death or MI during the phase of medical management, followed by a second gain at the time of percutaneous procedure, with the benefit maintained over a longer-term follow-up period.110Interventions were performed in 98% of patients in the CAPTURE trial, as mandated by the study protocol, in 30% of patients in the PRISM-PLUS trial, as clinically indicated, and in 13% of patients in the PURSUIT trial, as clinically indicated. Major bleeding by the TIMI criteria occurred in 0.8% of heparin-treated patients and in 1.4% of combination therapy-treated patients (p = 0.23) in the PRISM-PLUS trial, and occurred in 9.3% of the placebo recipients and 10.6% (p = 0.02) of the eptifibatide-treated patients in the PURSUIT trial. In the PRISM trial, using tirofiban without heparin, the rates of bleeding were 0.4% with heparin and with tirofiban. An economic assessment of the impact of using tirofiban in the practice pattern in Switzerland revealed a significant cost saving.111

Anticoagulant-Aspirin Combination Therapy
Parenteral Anticoagulants:

The results of trials and meta-analyses of heparin98100,112114 and of the combination of heparin with aspirin98,101,115117 were reviewed in the last consensus. Table 6 summarizes the results of trials that have compared unfractionated heparin plus aspirin to aspirin alone, complemented by the results of more recent trials that have compared a low-molecular-weight preparation plus aspirin to aspirin alone.119 The first LMWH trial randomized 219 patients to the open-labeled administration of a combination of aspirin plus nadroparin, aspirin alone, or the combination of unfractionated heparin plus aspirin.118The rates of ischemic events during hospitalization were 0%, 9.6%, and 5.7%, respectively, and the rates of nonfatal MI during hospitalization were 0%, 6%, and 9.5%, respectively. The Fragmin during Instability in Coronary Artery Disease (FRISC) trial compared the combination of dalteparin-aspirin administered for 35 to 45 days with aspirin alone in 1,506 patients.119 The rate of death and MI was reduced at 6 days from 4.8 to 1.8% (risk ratio, 0.37; 95% CI, 0.20 to 0.68; p = 0.001), and the need for revascularization was reduced from 1.2 to 0.4%. Although an excess number of events was observed when the dose of dalteparin was reduced from twice to once a day after 6 days, the benefit remained statistically significant after 40 days.

The LMWHs present distinct advantages over unfractionated heparin.120 They bind plasma proteins and endothelial cells less avidly than does unfractionated heparin, resulting in a more predictable degree of anticoagulation, and it can reliably and conveniently be administered SC once or twice a day with no need for monitoring. The LMWHs also produce fewer platelet agonist effects and are less often associated with heparin-induced thrombocytopenia. These advantages have been well-demonstrated in patients with venous disease and are now emerging in those with unstable angina.

Four large trials have directly compared an LMWH to unfractionated heparin; the results are summarized in Table 7 .121124 The Fragmin in Coronary Artery Disease (FRIC) study compared dalteparin (120 mg/kg every 12 h) with unfractionated heparin in 1,482 patients. No statistically significant differences between the two drug regimens were shown, with a rate of death or MI at day 6 of 9.3% among dalteparin-treated patients and 7.6% among unfractionated heparin-treated patients (difference NS).121In this trial, patients were rerandomized to double-blind administration of dalteparin (7,500 IU once daily) or placebo from day 6 to day 45. No statistically significant differences were observed between the two groups, but there was a trend to higher mortality with dalteparin treatment (dalteparin, 1.5%; placebo, 0.4%; 95% CI, 0.09 to 0.99; p = 0.057). The Enoxaparin in Unstable Angina and Non-Q-wave Myocardial Infarction (ESSENCE) trial randomized 3,171 patients to treatment with enoxaparin at doses of 1 mg/kg SC bid or to an IV infusion of unfractionated heparin for a minimum of 48 h and a maximum of 8 days.122The primary composite end point of death, MI, or recurrent angina at 14 days was reduced from 19.8 to 16.6% (p = 0.019) with enoxaparin, at 48 h it was reduced from 7.4 to 6.2% (p = 0.176), and at 30 days it was reduced from 23.3 to 19.8% (p = 0.016). The need for a revascularization procedure was reduced at 30 days from 32.2 to 27.0% (p = 0.001). The overall incidence of bleeding at 30 days was significantly higher with enoxaparin. However, major bleeding occurred in only 6.5% of enoxaparin-treated patients and in 7.0% of unfractionated heparin patients. In the TIMI-11B trial, 3,910 patients with unstable angina or non-Q-wave MI were randomized to treatment with IV unfractionated heparin for ≥ 3 days (median, 3.0 days) or to uninterrupted SC enoxaparin during the acute phase (median, 4.6 days), and, subsequently, to day 43.123The first dose of enoxaparin was administered IV. The primary end point of death, MI, or urgent revascularization at 8 days occurred in 14.5% of patients in the unfractionated heparin group and in 12.4% of patients in the enoxaparin group (p = 0.048), and by 43 days, in 19.7% and 17.3%, respectively (p = 0.048). The frequency of the secondary end point of death or MI was 2.1% and 1.7%, respectively, (p = 0.29) at 48 h, 5.9% and 4.6%, respectively, (p = 0.07) at 8 days, and 8.9% and 7.9%, respectively, (p = 0.28) at 43 days. The two survival curves remained parallel during the outpatient phase of treatment, suggesting no further treatment benefit of an additional 35 days of enoxaparin treatment. The rate of major hemorrhage was similar during the hospitalization period (unfractionated heparin, 1.0%; enoxaparin, 1.5%), and the rate of minor hemorrhage was higher with enoxaparin (2.5%; unfractionated heparin, 9.1%; p < 0.001). Outpatient treatment, compared with inpatient treatment, was associated with significantly more major bleeding (2.9% vs 1.5%, respectively; p = 0.02) and minor bleeding (19.3% vs 5.2%, respectively; p < 0.001). The Fraxiparine in Ischemic Syndrome (FRAXIS) trial involved 3,468 patients randomized to the unfractionated heparin group for 6 ± 2 days, SC nadroparin 86 anti-Xa IU/kg twice daily for 6 ± 2 days, or IV nadroparin 86 anti-Xa IU/kg followed by twice daily SC injections of 86 anti-Xa IU/kg for 14 days.124 No statistically significant differences were observed among the three treatment regimens with respect to the primary outcomes of cardiac death, MI, refractory angina, or recurrence of unstable angina at day 14 with rates of 18.1%, 17.8%, and 20.0%, respectively. The rates of death or MI at day 6 were 3.1%, 3.1%, 2.9%, respectively; at day 14, 4.5%, 5.3%, and 4.9%, respectively; and at 6 months, 7.9%, 8.8%, and 8.9%, respectively. Major bleeding during the first 6 days of treatment occurred in 1.0% of the unfractionated heparin group and in 1.0% of the two nadroparin groups combined. There was an increased risk of major hemorrhage with nadroparin administered for 14 days compared with unfractionated heparin (3.5% vs 1.6%, respectively; p = 0.0035).

A meta-analysis125 of the results of the ESSENCE and TIMI-11B trials with a total of 7,081 patients with similar entry criteria and similar doses of enoxaparin (except for an initial IV dose of enoxaparin and a longer period of administration in TIMI-11B) showed a 20% reduction in the risk of any ischemic event, which was statistically significant at 48 h and was maintained to 43 days. There was a statistically significant reduction in the rate of death or MI of 20% at 48 h (p = 0.24), of 23% at 8 days (p = 0.02), of 21% at 14 days (p = 0.02), and of 18% at 43 days (p = 0.02).125A meta-analysis of all trials that have compared any LMWH preparation to unfractionated heparin is presented in Table 7 . The end point of death, MI, or an ischemic event was reduced by 10% early (p = 0.02) and late (p = 0.15), and the end point of death or MI was reduced by 10% early (p = 0.19) and 9% late (p = 0.20). The reasons for the different results between enoxaparin, on one hand, and dalteparin and fraxiparine, on the other hand, are speculative. They could be related to different pharmacokinetic and pharmacodynamic properties of the drugs in relation to molecular weights or to different ratios of inhibition of factor Xa vs thrombin. Other possible explanations could be different study designs and different approaches to titration and monitoring of unfractionated heparin, resulting in varying intensity and stability of anticoagulation. Whatever the reasons for the differences, overall the trials have shown that LMWHs are associated with as favorable if not better results than unfractionated heparin. Given these results, the practical convenience of use, and the less frequent adverse platelet effects, LMWHs appear to be a better choice in patients with unstable angina than unfractionated heparin. An economic assessment of the ESSENCE results showed significant cost saving in US hospitals, but not across those in other countries, with the cost saving mainly attributable to fewer cardiac catheterization procedures.126 The administration of LMWH after hospital discharge does not seem warranted at the present time since the three trials (FRIC, TIMI 11B, and FRAXIS) that evaluated the strategy have not shown a trend to benefit, while the bleeding risk was significantly increased. These studies had some design limitations, however.

The FRISC II study used a different approach, as follows: 2,267 patients were randomized to double-blind administration of placebo or dalteparin twice daily for 3 months after a course of open administration of dalteparin to all patients for at least 5 days during the acute phase.127 Patients also were randomized in a factorial design to an invasive or noninvasive management strategy. A significant decrease in the composite end point of death or MI was observed at 30 days with dalteparin treatment (dalteparin, 3.1%; placebo, 5.9%; risk ratio, 0.53; 95% CI, 0.35 to 0.80; p = 0.002); the decrease was not statistically significant at 3 months (dalteparin, 6.7%; placebo, 8.0%; risk ratio, 0.81; 95% CI, 0.60 to 1.10; p = 0.17). The rate of the composite of death, MI, or revascularization was, however, significantly less at 3 months (dalteparin, 29.1; placebo, 33.4%; RR, 0.87; 95% CI, 0.77 to 0.99; p = 0.031), but not at 6 months. Patients who underwent early revascularization had no benefit from the long-term administration of dalteparin.

Previously, LMWH was replaced by unfractionated heparin before percutaneous interventions. Experience is now being gained with the LMWH during percutaneous interventions and with the concomitant administration of GPIIb/III antagonists and a thrombolytic agent. A pilot study with tirofiban and enoxaparin has suggested that the combination could be advantageous in providing more reproducible inhibition of platelet aggregation and no greater prolongation in bleeding time.128 Handling the LMWH when urgent surgery is needed remains a problem since the anticoagulation is not as readily reversed with protamine as it is with unfractionated heparin.

Oral Anticoagulants:

Oral anticoagulants have been continued following IV heparin in a few studies with the goal preventing reactivation of the disease and extending the benefit of therapy. This reactivation has been documented following the discontinuation of treatment with heparin,117 of LMWH,119 and of direct thrombin inhibitors.129 Three small studies have reported favorable results with oral anticoagulants. In the first study, performed with 102 patients, the total ischemic event rate was reduced after 6 months by 65% (p < 0.05).113 In the Antithrombotic Therapy in Acute Coronary Syndromes trial,116 involving 214 patients, there was a significant reduction after 2 weeks in the combined rate of death, MI, and recurrent ischemia from 27.5 to 10.5% (p = 0.004) with an INR prolonged to 2.0 to 2.5. The survival curves showed that most of the benefit occurred early during heparin therapy. The Organization to Assess Strategies for Ischemic Syndromes (OASIS)130pilot study of hirudin vs heparin compared a fixed-dose regimen of warfarin, 3 mg, in 309 patients, to a moderate dose titrated to an INR of 2 to 2.5 in 197 patients for 6 months. Aspirin was recommended for all patients. The low-intensity fixed dose of warfarin had no benefit. Moderate-intensity warfarin reduced the risk of death, MI, or refractory angina by 58% (p = 0.08) and reduced the need for rehospitalization for patients with unstable angina by 58% (p = 0.03).The OASIS 2 trial has evaluated this moderate-intensity warfarin on a larger scale. Preliminary results showed no benefit in the overall population of patients randomized, but in countries with stricter adherence to the protocol, there was an observed benefit. A small angiographic study of 57 patients randomized to treatment with aspirin and placebo or aspirin plus warfarin to an INR of 2 to 2.5 reported less progression in the treated group after 10 weeks in the severity of the obstruction at the site of the culprit lesion (patients, 4%; control subjects, 33%).131

Direct Thrombin Inhibitors

The direct thrombin inhibitors are potent anticoagulants that provide highly predictable inhibition of thrombin. They do not require a cofactor for their effect and have no known circulating inhibitors.132 Hirudin, the prototype of this class of agents, binds both the catalytic and anion-binding sites of thrombin. D-Phe-Pro-ArgCH2Cl, and synthetic small peptides like inogatran and argatroban, an arginine derivative, inhibit the catalytic site of thrombin. The anion-binding exosite site on thrombin can be occupied by hirugen and by a DNA aptamer. Bivalirudin (Hirulog; Biogen; Cambridge, MA) is a 20-amino acid peptide inhibiting the two active sites but with lower affinity than hirudin.132Large trials have been performed with hirudin and with bivalirudin in patients undergoing coronary angioplasty,133134 with hirudin and argatroban as adjunctive therapy to thrombolysis in patients who have experienced AMIs,129,135136 and with hirudin and inogatran in patients with unstable angina.129,137138 These trials, in general, have shown drug efficacy during treatment but have failed to be positive for their primary end point past the treatment phase. A major limitation encountered in the investigation of hirudin has been the bleeding complications, which lead to the discontinuation or reorientation of therapy with lower doses of three major trials.139141 The TIMI-9A139and the Global Use of Strategies to Open Occluded Coronary Arteries (GUSTO) IIA140 studies tested a 0.6-mg/kg bolus dose and 0.2-mg/kg/h infusion in patients with acute coronary syndromes without ST-segment elevation or as adjunctive therapy to thrombolysis. These two trials were discontinued early because of excess bleeding and resumed after reducing the dose of hirudin to a 0.1-mg/kg bolus and 0.1-mg/kg/h infusion.129,135 The OASIS-2 trial used a dose of 0.4-mg/kg bolus followed by 0.15-mg/kg/h infusion following favorable results observed with this regimen in the OASIS pilot study.142 The TIMI-9B trial showed no benefit of hirudin over heparin used as adjunctive therapy to tissue plasminogen activator or SK; death or MI at 30 days occurred in 9.7% of the hirudin-treated patients and 9.5% of the heparin-treated patients, and the primary composite end point of death, MI, CHF, or shock occurred in 12.9% and 11.9% of patients, respectively.135 In GUSTO IIB, the risk of death or MI at 30 days was reduced from 9.8% with heparin to 8.9% with hirudin (OR, 0.89; 95% CI, 0.79 to 1.0; p = 0.6).129 The reduction in the OR was of the same magnitude in patients randomized with or without ST-segment elevation. Of interest, the benefit of hirudin during the first 24 h was striking, with a 36% reduction in the event rate from 2.1 to 1.3% (p = 0.001) The OASIS-2 trial enrolled 10,241 patients with unstable angina or non-Q-wave MI.137 The primary end point of death or MI at 7 days was reduced by 16%, a difference that failed to reach statistical significance (p = 0.08). The inclusion of refractory angina in this end point resulted in a statistically significant difference (5.6% vs 6.7%, respectively; RR, 0.82; p = 0.01). In this trial, hirudin was more effective than heparin during the infusion period with a reduction at 72 h of 24% in the rate of death or MI (p = 0.04). A meta-analysis of the OASIS pilot study, OASIS-2, GUSTO IIB, and TIMI 9B results showed a highly statistically significant reduction of 28% in the risk of death or MI at 72 h (p = 0.0002) and of 17% at 7 days (p = 0.004) in patients with no ST-segment elevation; the reduction of 10% at 30 days was of borderline significance (p = 0.057). The trend to a benefit observed in patients receiving a thrombolytic agent was not statistically significant at any time points.137 No excess of major bleeding was observed with hirudin in the TIMI 9B and GUSTO IIB trials, whereas OASIS-2 reported a statistically significant excess (1.2% vs 0.7%, respectively; p = 0.01).

Trials with direct thrombin inhibitors other than hirudin observing patients with unstable angina have not conclusively shown a benefit. Bivalirudin was evaluated in 3,098 patients with unstable angina undergoing angioplasty for unstable angina or postinfarction angina.134 Bivalirudin did not reduce the primary end point of death, MI, abrupt vessel closure, or rapid clinical deterioration of cardiac origin during hospitalization, with observed rates of 11.4% compared with 12.2% with heparin. The subset of patients with postinfarction angina, however, benefited from the drug (9.1% vs 14.2%; p < 0.001), and there were fewer episodes of major hemorrhaging with bivalirudin (3.8% vs 9.8%, respectively; p < 0.001). A double-blind, randomized dose-finding study of inogatran in patients with unstable angina and non-Q-wave MI showed more ischemic events and a higher rate of death or MI with inogatran at the low-dose, medium-dose, and high-dose levels than with heparin.138

There have been four large trials of ASA administration to men free of a history of previous major vascular events (MI or stroke) (Table 8).90,143144,147 The Physicians’ Health Study142 was a double-blind, placebo-controlled, randomized trial designed to test two primary prevention hypotheses in a population of physicians free of MI, stroke, TIA, cancer, and current liver or renal disease, peptic ulcer, or gout. It was postulated that ASA might decrease mortality from cardiovascular disease and that beta-carotene might decrease cancer incidence. There were 22,071 male US physicians, aged 40 to 80 years, randomly allocated to treatment with low-dose ASA (eg, Bufferin; Bristol-Myers Squibb; New York, NY), 325 mg every other day, or placebo plus beta-carotene, 50 mg every other day, or placebo according to a 2 × 2 factorial design. The ASA component was terminated in 1998 at the recommendation of the Data and Safety Monitoring Board because of a clear reduction of MIs, a low likelihood of detecting a benefit of ASA on cardiovascular mortality before the year 2000, and the high prevalence of ASA use among participants following the occurrence of a nonfatal vascular event.

The principal outcome of cardiovascular death occurred at a rate of only 15% of that expected for a general population of white men with the same age distribution over a similar period and was not different between ASA (0.23% per year) and placebo (0.24% per year). The total death rate also was not different (ASA, 0.4% per year; placebo, 0.42% per year). There was a striking reduction in the rates of MI with ASA (0.26% per year) vs placebo (0.44% per year; risk reduction, 44%; p < 0.00001). The observed overall stroke rate was higher with aspirin (0.22% per year) vs placebo (0.18% per year; p = 0.15). The rate of hemorrhagic stroke was higher with aspirin (0.04% per year) vs placebo (0.02% per year; p = 0.06). The combined outcome of “important vascular events” (nonfatal MI, nonfatal stroke, and death from a cardiovascular cause) was significantly reduced in the ASA group (0.56% per year) vs the placebo group (0.68% per year; RRR 18%; p = 0.01).

The British Doctors’ Study144 was an open-label, randomly allocated trial of ASA, 500 mg daily, vs aspirin avoidance (2:1 ratio of ASA vs avoidance) that was conducted among British male physicians who had no history of stroke, definite MI, or peptic ulcer. There were 5,139 male physicians recruited and observed for up to 6 years. Vascular death rates, including that of sudden death from unknown cause, and those of peptic ulcer and gastric hemorrhage were lower with ASA (0.79% per year) vs no-ASA (ie, 0.84% per year; RRR 6%; p = NS).

Total mortality was not significantly less with ASA vs no-ASA (1.44% per year vs 1.6% per year; p = NS), and there were not fewer confirmed MIs (ASA, 0.9% per year; no-ASA, 0.93% per year; p = NS). Although there were significantly fewer confirmed TIAs in the ASA group (ASA, 0.16% per year; no-ASA, 0.28% per year; 2p < 0.05), there were more confirmed strokes in the ASA group (ASA, 0.32% per year; no-ASA, 0.29% per year; p = NS) and considerably more disabling strokes in the ASA group (ASA, 0.19% per year; no-ASA, 0.07% per year; risk ratio, 2.58; 2p < 0.05).

An overview of the two trials confirmed a reduced frequency of nonfatal AMI by ASA therapy (risk reduction, 32%; p < 0.0001), although there was heterogeneity of the observations (p = 0.035).145 The composite outcome of “any vascular event” was significantly reduced (RRR, 13%; p < 0.05), and there was a trend toward increased nonfatal stroke with aspirin therapy (risk increase, 18%; p = NS). There was no statistically significant reduction in total cardiovascular death or all-cause mortality in the overview.

The TPT90 was similar to the two previous trials in recruiting men who had not so far experienced major, clinically manifest episodes of ischemic heart disease (IHD) but differed in selecting those at increased risk. TPT recruited 5,499 men aged between 45 and 69 years at entry through 108 general practices in the United Kingdom. Those considered eligible fell in the top 20% of a risk score distribution based on smoking, family history, body mass index, BP, serum cholesterol level, plasma fibrinogen level, and factor VII activity, each weighted according to their associations with IHD in the first Northwick Park Heart Study.146 Of those patients who were eligible, 52% entered the trial. The two regimens evaluated consisted of low-intensity oral anticoagulation to an INR of about 1.5 with warfarin and a controlled-release 75-mg formulation of aspirin. The design was factorial with the following four treatment groups: active warfarin/active aspirin (WA); active warfarin/placebo aspirin (W); placebo warfarin/active aspirin; and placebo warfarin/placebo aspirin.

The mean warfarin dose required was 4.1 mg daily (range, 0.5 to 12.5 mg). There were 410 events of IHD (fatal, 142; nonfatal, 268). The main effect of warfarin (ie, comparing WA and W vs placebo warfarin/active aspirin and placebo warfarin/placebo aspirin) was a reduction in all IHD events of 21% (p = 0.02), chiefly due to a 39% reduction in fatal events (p = 0.003), so that warfarin reduced the death rate from all causes by 17% (p = 0.04). The main effect of aspirin (WA and placebo warfarin/active aspirin vs W and placebo warfarin/placebo aspirin) was a reduction in all IHD events of 20%, which was almost entirely due to a 32% reduction (p = 0.004) in nonfatal events. Recent analyses have suggested, first, a strong interaction between recruitment systolic BP and the treatment effect of ASA, and that those with BP levels ≤ 130 mm Hg derived considerably more benefit than those with higher pressures, there being neither a beneficial nor a harmful effect in those patients with BPs ≥ 145 mm Hg. Second, although it was less clear, there appears to have been a significant excess of fatal coronary events in men aged ≥ 65 years at entry. This could account for the 12% higher overall mortality rate from coronary events. In the individual treatment groups, the absolute reductions in all IHD events due to absolute reductions of IHD events compared with placebo were the following: warfarin, 2.6 events per 1,000 person-years; aspirin, 2.3 events per 1,000 person-years; and warfarin/aspirin, 4 events per 1,000 person-years. Neither W nor placebo warfarin/active aspirin alone affected the incidence of all strokes, although WA increased hemorrhagic strokes (p = 0.009). Of the 10 hemorrhagic strokes that occurred, 7 were in the WA group, and the mean systolic BP of these men at trial entry was 158 mm Hg, compared with 146 mm Hg in those experiencing other strokes and 135 mm Hg in those who did not have strokes. Major noncerebral bleeding episodes were about twice as frequent in the active treatment groups as in the placebo-warfarin-plus-placebo-aspirin group, but the differences were NS and there was no significant difference in frequency among the three active treatment groups (WA, W, and placebo warfarin/active aspirin). Less serious bleeding occurred more frequently in the WA group than in the other groups.

As its name implies, the Hypertension Optimal Treatment (HOT) trial was principally concerned with the management of hypertension, more specifically to assess the optimum target diastolic BP in those with hypertension.147 However, it also randomized participants to treatment with aspirin or placebo. A total of 19,193 subjects from 26 countries with ages between 50 and 80 years (mean, 61.5 years) with diastolic BPs between 100 and 115 mm Hg (average, 105 mm Hg) were randomly assigned a target BP and also were randomly assigned to treatment with a 75-mg dose of aspirin or placebo. Irregularities in one center led to the exclusion of 403 patients early on. The average follow-up time was 3.8 years (range, 3.3 to 4.9 years), giving a total of 71,051 patient-years.

Participants were randomly assigned to one of the following three diastolic BP target groups: ≤ 90 mm Hg; ≤ 85 mm Hg; or ≤ 80 mm Hg. Antihypertensive therapy with felodipine, 5 mg once daily, was administered to all participants. Additional therapy and dose increments were with angiotensin-converting enzyme inhibitors or β-blockers with the possibility of also adding a diuretic agent. Aspirin treatment was with 75 mg aspirin daily or an identical placebo tablet.

Major cardiovascular events were defined as all MIs (fatal and nonfatal), all strokes (fatal and nonfatal), and all other cardiovascular deaths. Silent MI was documented by ECGs at randomization and at the final visit.

In summary, the BP-lowering (and main) component of the trial showed reductions in diastolic BP of 20.3 mm Hg, 22.3 mm Hg, and 24.3 mm Hg, respectively, in the target groups of ≤ 90 mm Hg, ≥ 85 mm Hg, and ≤ 80 mm Hg. The lowest incidence of major cardiovascular events occurred at a mean achieved diastolic BP of 82.6 mm Hg, and the lowest risk of cardiovascular mortality occurred at 86.5 mm Hg. Further reduction below these BPs was safe.

There were 209 episodes of MI, 82 in those assigned to ASA and 127 to placebo, representing a reduction of 36% (p = 0.002) and the prevention of 1.5 episodes per 1,000 person-years. There were 315 major cardiovascular events in those patients receiving ASA compared with 368 in the placebo group, a reduction that was just significant at a conventional level. However, there were no clear differences in cardiovascular mortality or total mortality. The number of strokes was almost identical in the two groups, including episodes of cerebral hemorrhage. However, nonfatal major bleeding and minor bleeding occurred more frequently in those receiving active treatment. The results of the HOT trial also suggest that the main beneficial effect of ASA is a reduction of the number of nonfatal MIs.

Participants in the HOT trial were at intermediate risk on account of their raised entry BP levels, and all received active pressure-lowering regimens. The rate of all MIs was reduced by ASA therapy from 3.6 to 2.1 events per 1,000 patient-years (relative reduction, 36%; absolute reduction, 1.5 events per 1,000 patient-years). There was no difference between the ASA-treated and the placebo groups in terms of the number of fatal hemorrhagic events, but there were 129 nonfatal major bleeding events in the ASA-treated group compared with 70 in the placebo group, the excess being mainly attributable to GI, nasal, and “other” episodes. There were 12 nonfatal cerebral bleeding events in each group. There were 156 minor bleeding episodes in the ASA-treated group compared with 87 in the placebo group, the main contribution to this excess being nasal bleeding.

ASA therapy reduced ischemic cardiac events in three of the four trials, the effect being most marked for nonfatal MI. Although there were trends to increased total stroke and hemorrhagic stroke with ASA in the first two trials (ie, the US Physicians Trial and the UK Doctors Trial), there was a trend toward a lower number of total strokes with ASA in the third trial (TPT) and virtually no difference in the fourth (HOT). A main distinguishing characteristic between the first two trials and the second two was the considerably lower dose of aspirin, 75 mg daily, in the latter two (TPT and HOT). In the absence of compelling evidence of greater benefit with higher doses, it seems reasonable to advise the use of ≤ 75 mg aspirin daily for primary prevention. There is a consistent failure in all four trials to show a reduction in all-cause mortality by ASA. In the US Physicians’ Trial,143 the risk of MI among men aged 40 to 49 years was only 0.1% per year (1 MI per year per 1,000 men), whereas among men aged 60 to 69 years, the rate of MI was 0.82% per year (8.2 MIs per year per 1,000 men). Among the older men, the absolute risk reduction with ASA was about 4.4 infarcts per year per 1,000 men treated. Similarly, the absolute benefits were greater among men with diabetes mellitus, with systolic or diastolic hypertension, who smoked cigarettes, and who had a lack of exercise. In the TPT,90 the patients enrolled were at substantially increased risk of cardiovascular events, the event rates in the placebo group were substantially higher than in the US Physicians’ Trial, and the absolute benefits of ASA were greater. Although even among the higher-risk patients there is no evidence for a reduction of all-cause mortality, the reduction of (mainly nonfatal) ischemic cardiac and ischemic cerebrovascular events probably outweighs the risk of hemorrhagic stroke, other bleeding, and GI side effects and may justify the prophylactic use of aspirin among men with no history of AMI, stroke, or TIA. However, a recent overview of ASA trials has raised the possibility that even in men at moderately increased risk, the benefit of treatment may not in fact outweigh the risk of cerebral hemorrhage.148149 The overview included results from secondary as well as primary prevention trials, but if the likelihood of cerebral hemorrhage is more or less constant regardless of the setting in which ASA is used, while absolute benefits will indeed vary depending on the context, there does seem to be uncertainty about the balance between benefit and risk. However, nearly all the trials contributing to the overview used > 75 mg ASA daily, so that, taking into account the findings on stroke in the TPT and the HOT trial (see previous section), 75 mg ASA daily may confer a clear net benefit in those patients at moderate risk as well as those at high risk. With the necessary caveats about subgroup analyses, the results of the TPT quite strongly suggest that the benefit of aspirin is greater the lower the patient’s BP, its use in those with higher pressures being neither beneficial nor harmful.150 It is also possible, however, that aspirin increases the number of fatal events of coronary heart disease in older men,150 although this observation requires confirmation or refutation in other trials.

Warfarin appears to have similar efficacy to ASA for the prevention of all IHD outcomes, but it is particularly effective in reducing fatal events, resulting in a statistically significant reduction in all-cause mortality (RRR, 17%; p = 0.04). A combination of warfarin and aspirin was particularly effective in reducing ischemic cardiac events. Although hemorrhagic stroke was increased by the warfarin/aspirin combination, this risk can probably be minimized by careful BP monitoring and effective antihypertensive therapy. Compared with placebo, the risk of noncerebral major bleeding was increased to a similar degree by aspirin, warfarin, and a combination of the two.

Among men at very low risk for cardiovascular events, the risk of cerebrovascular or other bleeding with either ASA or warfarin outweighs the benefit. As the risk of cardiovascular events increases, the potential benefits of ASA and warfarin increase, while the bleeding risk is likely to remain about constant and therapy, therefore, becomes justified. The lower cost, greater ease of use, and lower risk of ASA lead to its recommendation over warfarin. A combination of ASA and warfarin may be indicated among those individuals at particularly high risk for cardiovascular events. Lowering diastolic BP to ≤ 85 mm Hg should be a goal in any patient receiving ASA or warfarin for primary prevention.

The primary prevention trials of aspirin have not included women. To our knowledge, the only available data are from a large prospective cohort study of 28,678 US registered nurses, aged 34 to 65 years, who had not received diagnoses of coronary artery disease, stroke, and cancer at baseline.151Among women taking one to six aspirins per week, the age-adjusted RR of a first MI was 0.68 (p = 0.005). This benefit was confined to women ≥ 50 years (RR, 0.61; p = 0.002). There were trends toward fewer deaths from cardiovascular events (RR, 0.89; p = 0.56) and fewer important vascular events (RR, 0.85; p = 0.12), but there was no difference for the incidence of stroke (RR, 0.99). No benefits were observed among women taking more than six aspirins per week. As in any cohort trial, there are concerns that unanticipated and undocumented confounders may bias the conclusions. Such factors might explain the failure to observe a benefit among the women taking more than six aspirins per week, although a β error might also be possible. Overall, the findings support those of aspirin trials in men. New data should come from the Women’s Health Study.152

In other settings of vascular disease, there are trials that indicate that women benefit from aspirin therapy when the underlying problem is unstable angina97 or AMI.65 However, the lower age-matched risk of cardiac events for women compared with men suggests that there are likely to be fewer absolute benefits among women at a given age, particularly if there is no difference between men and women in bleeding episodes.

Of the 22,071 male physicians without a history of MI, stroke, or transient cerebral ischemia enrolled in the Physician’s Health Study, 333 had a history of exertional angina.153 During 60.2 months of follow-up, those randomly assigned to receive aspirin had a 70% RRR for MI compared with those assigned to receive placebo (p = 0.003). Controlling for other cardiovascular risk factors, the RRR was 87% (p < 0.001).

The Swedish Angina Pectoris Trial Group154 described 2,035 patients with stable angina randomly assigned to receive aspirin, 75 mg daily, or placebo for a median of 15 months. All patients received sotalol for control of angina. There was a 34% reduction of the primary outcome of MI and sudden death (CI, 24 to 49%; p = 0.003). Other vascular events and mortality were reduced by 22 to 32%. This was a secondary prevention study, with angina serving as a marker for coronary artery disease, just as prior MI, stroke, or TIA also serve as markers.

Anticoagulant Therapy When Thrombolytic Therapy Has Been Administered:

For all patients who have received rtPA or reteplase, we recommend administration of heparin according to the following regimen: (1) a bolus of about 60 U/kg IV to a maximum dose of 4,000 U at the initiation of rtPA infusion or a first bolus of rPA or TNK with an initial maintenance dose of 12 U/kg/h to a maximum dose of 1,000 U, with APTT at 1.5 to 2 times control, maintained for 48 h; and (2) a maintenance of the APTT at 1.5 to 2 times control beyond 48 h should be undertaken only in the presence of determinants of high risk of systemic or venous thromboembolism (eg, anterior Q-wave infarction, severe LV dysfunction, CHF, history of systemic or pulmonary embolus, 2D echocardiographic evidence of mural thrombosis, or atrial fibrillation). In such cases, the IV regimen may be sustained, or consideration may be given to SC administration (an initial dose of approximately 17,500 U every 12 h to maintain APTT at 1.5 to 2 times control), LMWH SC, or to conversion to warfarin therapy (target INR, 2.5; INR range, 2.0 to 3.0) for up to 3 months (grade 2A). For patients with atrial fibrillation, we recommend warfarin therapy (target INR, 2.5; INR range, 2.0 to 3.0) indefinitely (grade 1A).

For all patients who have received SK or APSAC, we recommend the administration of IV heparin only in the presence of determinants for a high risk of systemic or venous thromboembolism (eg, anterior AMI, CHF, previous embolus, or atrial fibrillation) and then according to the following regimen: (1) measure APTT when the indication emerges but not < 4 h after beginning SK or APSAC infusion; if more than two times control, repeat APTT as appropriate, and commence infusion of heparin when APTT is less than two times control and maintain APTT at 1.5 to 2 times control as long as the risk of thromboembolism is considered to be high; (2) after 48 h, consideration may be given to SC administration (initial dose approximately 17,500 U every 12 h to maintain APTT at 1.5 to 2 times control), LMWH SC, or for conversion to warfarin therapy (target INR, 2.5; INR range, 2.0 to 3.0) for up to 3 months (grade 2A). For patients with atrial fibrillation, we recommend warfarin therapy (target INR, 2.5; INR range, 2.0 to 3.0) indefinitely (grade 1A).

Anticoagulant Therapy When No Thrombolytic Therapy Has Been Given:

For patients at increased risk for systemic or pulmonary embolism because of anterior Q-wave infarction, severe LV dysfunction, CHF, history of systemic or pulmonary embolism on 2D echocardiographic evidence of mural thrombosis, we recommend the administration of heparin (about 75-U/kg bolus IV; initial maintenance dose, 1,000 to 1,200 U/h IV; APTT, 1.5 to 2 times control), followed by warfarin (target INR, 2.5; INR range, 2.0 to 3.0) for up to 3 months (grade 2A). For patients with atrial fibrillation, we recommend warfarin therapy (target INR, 2.5; INR range, 2.0 to 3.0) indefinitely (grade 1A).

We recommend that clinicians use not less than low-dose heparin therapy (ie, 7,500 U SC every 12 h) or LMWH until the patient is ambulatory for the prevention of venous thrombosis, unless there is a specific contraindication, in every patient with AMI (grade 1A).

Antiplatelet Therapy

  1. Aspirin doses of 75 to 162.5 mg have been shown to be effective in all indications, and this dose range is recommended. Since the onset of full antiplatelet activity is delayed with low doses, if a rapid response is required (ie, in patients with MI or stroke), a dose of 162.5 mg should be used.

  2. We recommend that all patients with AMIs receive non-enteric-coated aspirin to chew and swallow as soon as possible after the clinical impression of evolving AMI is formed, and whether or not thrombolytic therapy is to be given. Daily aspirin administered orally should be continued indefinitely (grade 1A).

  3. If the patient is to receive heparin, we recommend administering aspirin conjointly (grade 2A).

  4. If warfarin therapy is commenced, we recommend discontinuing aspirin therapy until the planned course of warfarin is complete. Aspirin therapy then should be restarted and maintained indefinitely. We recommend that clinicians not administer aspirin concurrently with warfarin, except in situations of very high embolic risk or previous failure of either therapy alone (grade 2C).

  5. When embolic risk is low, we recommend long-term aspirin therapy in preference to warfarin because of its simplicity, safety, and low cost (grade 2A).

  6. We recommend the use of long-term warfarin therapy in clinical settings of increased embolic risk for a duration of 1 to 3 months following anterior AMI, or AMI complicated by severe LV dysfunction, CHF, previous emboli, or 2D echocardiographic evidence of mural thrombosis atrial fibrillation (grade 2A). For patients with atrial fibrillation, we recommend warfarin therapy (target INR, 2.5; INR range, 2.0 to 3.0) indefinitely (grade 1A).

  7. We recommend that patients who have contraindications to aspirin should receive clopidogrel (75 mg/d) indefinitely (grade 1A).

  8. A further alternative for patients who have contraindications to aspirin is that clinicians give warfarin (target INR, 2.5). The increased complexity, risk, and cost of such therapy are concerns (grade 2A).

  9. Some patients with recurrent ischemic episodes following AMI may benefit from a combination of warfarin and aspirin. We recommend that clinicians offer treatment with low-dose aspirin (75 to 80 mg) and low-intensity warfarin to these patients (target INR, 1.5) (grade 2C).

  10. We recommend the use of aspirin rather than sulfinpyrazone for survivors of AMI because of the evidence for a benefit from aspirin, which is a less expensive agent with a simpler dose regimen, and because of more extensive evidence supporting its efficacy (grade 1C).

  11. We do not recommend that clinicians use dipyridamole alone (grade 2C) or in combination with aspirin (grade 2B) in survivors of AMI.

Unstable Angina
Antiplatelet Agents:

Aspirin doses of 75 to 162.5 mg have been shown to be effective in all indications.

In patients with unstable angina, we recommend the administration of non-enteric-coated aspirin to chew and swallow as soon as possible after the clinical impression of unstable angina is formed. Aspirin administered orally should be continued indefinitely (all grade 1A).


We recommend that patients with unstable angina, who have aspirin allergy or intolerance, receive clopidogrel (75 mg daily) (grade 1C), ticlopidine (250 mg bid) (grade 1A), triflusal (in countries where it is available) (grade 1A), or warfarin (target INR, 2.5) for several months (grade 2C).

We recommend that clinicians not administer sulfinpyrazone to patients with unstable angina (grade 1C).

We recommend the administration of IV tirofiban or eptifibatide, in addition to aspirin and heparin, to patients with continuing ischemia or other high-risk features. The indication is strengthened by the detection of elevated levels of troponin T or troponin I. The infusion should continue for 48 to 72 h, or until percutaneous intervention (grade 1A).

We recommend the administration of abciximab for 12 to 24 h in patients who will undergo percutaneous intervention within the following 24 h (grade 1A).

Anticoagulant Aspirin Combination Therapy:

In patients hospitalized with unstable angina, we recommend, in addition to aspirin therapy, commencement of therapy with IV heparin (about 75 U/kg IV bolus, initial maintenance 1,250U/h IV, APTT 1.5 to 2 times control) or LMWH (dose regimens from trials). The therapy should be maintained for at least 48 h, or until the unstable pain pattern resolves with the present or more definitive therapy (grade 1A).

Direct Thrombin Inhibitors:

Although therapy with hirudin plus aspirin offers benefit over unfractionated heparin plus aspirin, in view of cost, hemorrhagic risk, and availability of competing agents, we recommend heparin as the agent of choice (grade 2A). We recommend hirudin over heparin for patients with a history of heparin-induced thrombocytopenia (grade 1C).

Primary Prevention

We do not recommend the routine use of aspirin for the primary prevention of coronary artery disease outcomes in individuals free of a history of AMI, stroke, or transient cerebral ischemic attack who are < 50 years of age (grade 2B).

For individuals free of a history of prior MI, stroke, or transient cerebral ischemic attack but with increasing levels of risk, there are data available for the efficacy of aspirin, warfarin, and the combination. Because of the increased complexity and costs of treatment with warfarin, and because of the greater likelihood of cerebral hemorrhage with the combination of aspirin and warfarin, the following recommendations are made for individuals at increasing risk of cardiovascular events.

  1. We recommend that aspirin be considered for men > 50 years of age who have at least one major risk factor for coronary artery disease and who are free of contraindications to aspirin (grade 2A).

  2. We recommend that aspirin be considered for women > 50 years of age who have at least one major risk factor for coronary artery disease (ie, cigarette smoking, hypertension, diabetes mellitus, high cholesterol level, and history of parental infarction) and who are free of contraindications to aspirin (grade 2C).

  3. We recommend that low-intensity warfarin therapy (target INR, 1.5) be considered as an alternate to aspirin for men at high risk of cardiovascular events in the prevention of those events and for reduction of all-cause mortality (grade 2A).

  4. We recommend that a combination of low-dose aspirin therapy (ie, 75 to 80 mg/d) and low-intensity warfarin therapy (target INR, 1.5) be considered as an alternative to aspirin or warfarin alone for men who are at very high risk of cardiovascular events for the prevention of these events and the reduction of all-cause mortality (grade 2A).

  5. Whenever antithrombotic therapy is prescribed for primary prevention, we recommend aggressive BP control (target diastolic BP, < 85 mm Hg) (grade 1C).

Chronic Coronary Artery Disease

  1. We recommend administering oral aspirin to all patients with stable angina indefinitely (grade 1A).

  2. We recommend that all patients with clinical or laboratory evidence of coronary artery disease receive oral aspirin indefinitely (grade 2C).

Abbreviations: ADP = adenosine diphosphate; AMI = acute myocardial infarction; APSAC = anistreplase; APTT = activated partial thromboplastin time; ASA = acetylsalicylic acid; ASPECT trial = Anticoagulation in the Secondary Prevention of Events in Coronary Thrombosis; CAPTURE trial = c7E3 Fab Antiplatelet Therapy in Unstable Refractory Ischemia; CHAMP study = Combined Hemotherapy and Mortality Prevention; CHF = congestive heart failure; CI = confidence interval; CK = creatine kinase; EF = ejection fraction; ESSENCE trial = Enoxaparin in Unstable Angina and Non-Q-Wave Myocardial Infarction; FRAXIS trial = Fraxiparine in Ischemic Syndrome; FRIC trial = Fragmin in Coronary Artery Disease; FRISC trial = Fragmin during Instability in Coronary Artery Disease; GP = glycoprotein; GUSTO trial = Global Use of Strategies to Open Occluded Coronary Arteries; HOT = Hypertension Optimal Treatment; IHD = ischemic heart disease; INR = international normalized ratio; ISIS = International Studies of Infarct Survival; LMWH = low-molecular-weight heparin; LV = left ventricle, ventricular; MI = myocardial infarction; NS = not significant; OASISstudy = Organization to Assess Strategies for Ischemic Syndromes; OR = odds ratio; PRISM trial = Platelet Receptor Inhibition for Ischemic Syndrome Management; PTCA = percutaneous transluminal coronary angioplasty; PURSUIT trial = Platelet IIb/IIIa in Unstable Angina Receptor Suppression Using Integrilin; RR = relative risk; RRR = relative risk reduction; rtPA = recombinant tissue-type plasminogen activator; SC = subcutaneous, subcutaneously; SCATI group = Studio sulla Calciparina nell’Angina e nella Thrombosi Ventriculare Nell’Infarto; SK = streptokinase; TIA = transient ischemic attack; TIMI = Thrombolysis in Myocardial Infarction; TPT = Thrombosis Prevention Trial; 2D = two-dimensional; W = active warfarin and placebo aspirin; WA = active warfarin and active aspirin; WARIS = Warfarin Reinfarction Study

Table Graphic Jump Location
Table 1. Outcomes of Heparin Therapy in Patients With Suspected AMI*

Values given as No. of events per 1,000 patients, unless otherwise indicated. Values in parentheses indicate that 93% of patients received a thrombolytic drug. Table was constructed from the overview of 26 randomized, unconfounded trials by Collins et al.3233

Table Graphic Jump Location
Table 2. Long-term Trials of Anticoagulation Following MI
Table Graphic Jump Location
Table 3. Trials of Aspirin vs Oral Anticoagulation Following AMI*

Anticoag = anticoagulant; ARR = absolute risk reduction; AFTER study = aspirin and anticoagulation following thrombolysis for myocardial infarction; CARS study = Coumadin, Aspirin Reinfarction.


One milligram of anticoagulant.


Log rank test.


Three milligrams of anticoagulant.

Table Graphic Jump Location
Table 4. Trials That Have Compared an Oral Antiplatelet Agent to Placebo in Patients With Unstable Angina*

Ticlo = ticlopidine; Sulf = sulfinpyrazone.


All trials: 95% CI for all-cause mortality, 14 to 54.


All trials: 95% CI for cardiac death/nonfatal MI, 32 to 54.


Intent-to-treat analysis.

Table Graphic Jump Location
Table 5. Trials With GPIIb/IIIa Antagonists in Patients With Unstable Angina*

PARAGON = Platelet IIb/IIIa Antagonism for the Reduction of Acute Coronary Syndrome Events in a Global Organization Network; PCI = percutaneous intervention.


All trials: 95% CI for death/MI at 30 d, 7 to 12.

Table Graphic Jump Location
Table 6. Trials That Have Compared the Combination Heparin-Plus-Aspirin to Aspirin Alone in Patients With Unstable Angina*

Combination = combination therapy with unfractionated heparin or LMWH.