On Defining a Myocardial Infarction

More than 25 years ago, the benefit of the routine treatment of patients with β-adrenergic blocking agents who experienced a myocardial infarction was established by the Norwegian Multicenter Study on Timolol After Myocardial Infarction and the Beta Blocker Heart Attack Trial (BHAT). At that time, the placebo mortality in BHAT was 8.9% compared with 6.6% in patients treated with propranolol and represented a 26% decrease in all cause mortality.

Since then, our targets for the successful treatment in patients experiencing an acute MI and more recently for an acute coronary syndrome have changed. Over the years, the mortality rate in acute MI in ACS has decreased to 2%–3% as a result of therapy with a variety of agents, including β-blockers.

At the same time, our diagnostic criteria for acute MI have changed. Patients with lesser degrees of myocardial necrosis can now be identified using sensitive biomarkers such as the troponin assays. As a result, the incidence of nonfatal MIs, repeat hospitalizations, and revascularization events has increased markedly.

In contemporary trials for the treatment of ACS, nonfatal MIs represent four to five times the number of mortality events. To adequately measure significant and modest treatment effects, current trials often require more than 10,000 patients. The primary therapeutic goal in these trials is a composite end point that includes both cardiac deaths and nonfatal MIs. Since a mortality benefit is difficult to demonstrate because of relatively few events, most of the benefit is measured by a decrease in nonfatal MIs. They are defined in part by the TIMI clinical definition but driven to an even greater extent by elevated troponin concentrations above the upper limit of normal. The degree of elevation is not reported. We are all aware of the uncertainty of this highly sensitive determination of cell death and its relative lack of specificity, particularly at the lower concentrations.

The importance of nonfatal MI frequency in determining clinical trial efficacy was dramatically shown recently in TRITON (Trial to Assess Improvement in Therapeutic Outcomes by Optimizing Platelet Inhibition With Prasugrel—Thrombolysis in Myocardial Infarction), which compared prasugrel to clopidogrel for the treatment of ACS patients undergoing percutaneous coronary intervention. In that study of 13,608 patients, there were 183 cardiovascular deaths and 1,095 nonfatal MIs. The combined end point was significantly reduced with prasugrel, compared with clopidogrel, determined mainly by a significant decrease in nonfatal MI with prasugrel. Yet prasugrel had no significant effect on mortality, compared with clopidogrel. In the trial prasugrel was, however, associated with a significant increase in life-threatening and fatal bleeding events. It is therefore clear from these data that in TRITON we are using a potent agent with significant bleeding risks for the prevention of recurrent nonfatal MIs, the nature and importance of which is unclear.

A subgroup analysis of TRITON suggested that patients with certain characteristics were at an increased risk of bleeding if given prasugrel: those who had a previous stroke, those who were aged over 75, or those who were underweight. Although this analysis might provide comfort to some physicians, it does not provide insight into what long-term benefit can be achieved by preventing a nonfatal MI.

A recent market survey of 85 interventional and clinical cardiologists reported in Forbes.com surprisingly showed that 36% of their patients would receive prasugrel following angioplasty. Even more striking was the fact that some physicians might use prasugrel for the long-term treatment of ACS patients who did not undergo percutaneous coronary intervention (www.forbes.com/healthcare/2007/11/29/prasugrel-plavix-lilly-biz-healthcare-cx_mh_1129lilly.html

It is clear that patients with suspected ACS with elevated troponin represent a very heterogeneous group. The need to know more about this group of patients is emphasized in the recent consensus document establishing a universal definition of MI, which focuses particularly on the uncertain definition of nonfatal MI in randomized clinical trials. The document emphasizes that we know little about the long term outcome in this mixed patient population. The new definition provides a classification of nonfatal MI for future clinical trials based on the degree of troponin elevation and the mechanism of infarction.

Before we proceed in the direction of prevention of all nonfatal MIs with drugs that have major downside risks, we need to know more about who will benefit. It is imperative that future clinical trials better define nonfatal MI and provide information about the therapeutic benefit of preventing that event.

More than 25 years ago, the benefit of the routine treatment of patients with β-adrenergic blocking agents who experienced a myocardial infarction was established by the Norwegian Multicenter Study on Timolol After Myocardial Infarction and the Beta Blocker Heart Attack Trial (BHAT). At that time, the placebo mortality in BHAT was 8.9% compared with 6.6% in patients treated with propranolol and represented a 26% decrease in all cause mortality.

Since then, our targets for the successful treatment in patients experiencing an acute MI and more recently for an acute coronary syndrome have changed. Over the years, the mortality rate in acute MI in ACS has decreased to 2%–3% as a result of therapy with a variety of agents, including β-blockers.

At the same time, our diagnostic criteria for acute MI have changed. Patients with lesser degrees of myocardial necrosis can now be identified using sensitive biomarkers such as the troponin assays. As a result, the incidence of nonfatal MIs, repeat hospitalizations, and revascularization events has increased markedly.

In contemporary trials for the treatment of ACS, nonfatal MIs represent four to five times the number of mortality events. To adequately measure significant and modest treatment effects, current trials often require more than 10,000 patients. The primary therapeutic goal in these trials is a composite end point that includes both cardiac deaths and nonfatal MIs. Since a mortality benefit is difficult to demonstrate because of relatively few events, most of the benefit is measured by a decrease in nonfatal MIs. They are defined in part by the TIMI clinical definition but driven to an even greater extent by elevated troponin concentrations above the upper limit of normal. The degree of elevation is not reported. We are all aware of the uncertainty of this highly sensitive determination of cell death and its relative lack of specificity, particularly at the lower concentrations.

The importance of nonfatal MI frequency in determining clinical trial efficacy was dramatically shown recently in TRITON (Trial to Assess Improvement in Therapeutic Outcomes by Optimizing Platelet Inhibition With Prasugrel—Thrombolysis in Myocardial Infarction), which compared prasugrel to clopidogrel for the treatment of ACS patients undergoing percutaneous coronary intervention. In that study of 13,608 patients, there were 183 cardiovascular deaths and 1,095 nonfatal MIs. The combined end point was significantly reduced with prasugrel, compared with clopidogrel, determined mainly by a significant decrease in nonfatal MI with prasugrel. Yet prasugrel had no significant effect on mortality, compared with clopidogrel. In the trial prasugrel was, however, associated with a significant increase in life-threatening and fatal bleeding events. It is therefore clear from these data that in TRITON we are using a potent agent with significant bleeding risks for the prevention of recurrent nonfatal MIs, the nature and importance of which is unclear.

A subgroup analysis of TRITON suggested that patients with certain characteristics were at an increased risk of bleeding if given prasugrel: those who had a previous stroke, those who were aged over 75, or those who were underweight. Although this analysis might provide comfort to some physicians, it does not provide insight into what long-term benefit can be achieved by preventing a nonfatal MI.

A recent market survey of 85 interventional and clinical cardiologists reported in Forbes.com surprisingly showed that 36% of their patients would receive prasugrel following angioplasty. Even more striking was the fact that some physicians might use prasugrel for the long-term treatment of ACS patients who did not undergo percutaneous coronary intervention (www.forbes.com/healthcare/2007/11/29/prasugrel-plavix-lilly-biz-healthcare-cx_mh_1129lilly.html

It is clear that patients with suspected ACS with elevated troponin represent a very heterogeneous group. The need to know more about this group of patients is emphasized in the recent consensus document establishing a universal definition of MI, which focuses particularly on the uncertain definition of nonfatal MI in randomized clinical trials. The document emphasizes that we know little about the long term outcome in this mixed patient population. The new definition provides a classification of nonfatal MI for future clinical trials based on the degree of troponin elevation and the mechanism of infarction.

Before we proceed in the direction of prevention of all nonfatal MIs with drugs that have major downside risks, we need to know more about who will benefit. It is imperative that future clinical trials better define nonfatal MI and provide information about the therapeutic benefit of preventing that event.

More than 25 years ago, the benefit of the routine treatment of patients with β-adrenergic blocking agents who experienced a myocardial infarction was established by the Norwegian Multicenter Study on Timolol After Myocardial Infarction and the Beta Blocker Heart Attack Trial (BHAT). At that time, the placebo mortality in BHAT was 8.9% compared with 6.6% in patients treated with propranolol and represented a 26% decrease in all cause mortality.

Since then, our targets for the successful treatment in patients experiencing an acute MI and more recently for an acute coronary syndrome have changed. Over the years, the mortality rate in acute MI in ACS has decreased to 2%–3% as a result of therapy with a variety of agents, including β-blockers.

At the same time, our diagnostic criteria for acute MI have changed. Patients with lesser degrees of myocardial necrosis can now be identified using sensitive biomarkers such as the troponin assays. As a result, the incidence of nonfatal MIs, repeat hospitalizations, and revascularization events has increased markedly.

In contemporary trials for the treatment of ACS, nonfatal MIs represent four to five times the number of mortality events. To adequately measure significant and modest treatment effects, current trials often require more than 10,000 patients. The primary therapeutic goal in these trials is a composite end point that includes both cardiac deaths and nonfatal MIs. Since a mortality benefit is difficult to demonstrate because of relatively few events, most of the benefit is measured by a decrease in nonfatal MIs. They are defined in part by the TIMI clinical definition but driven to an even greater extent by elevated troponin concentrations above the upper limit of normal. The degree of elevation is not reported. We are all aware of the uncertainty of this highly sensitive determination of cell death and its relative lack of specificity, particularly at the lower concentrations.

The importance of nonfatal MI frequency in determining clinical trial efficacy was dramatically shown recently in TRITON (Trial to Assess Improvement in Therapeutic Outcomes by Optimizing Platelet Inhibition With Prasugrel—Thrombolysis in Myocardial Infarction), which compared prasugrel to clopidogrel for the treatment of ACS patients undergoing percutaneous coronary intervention. In that study of 13,608 patients, there were 183 cardiovascular deaths and 1,095 nonfatal MIs. The combined end point was significantly reduced with prasugrel, compared with clopidogrel, determined mainly by a significant decrease in nonfatal MI with prasugrel. Yet prasugrel had no significant effect on mortality, compared with clopidogrel. In the trial prasugrel was, however, associated with a significant increase in life-threatening and fatal bleeding events. It is therefore clear from these data that in TRITON we are using a potent agent with significant bleeding risks for the prevention of recurrent nonfatal MIs, the nature and importance of which is unclear.

A subgroup analysis of TRITON suggested that patients with certain characteristics were at an increased risk of bleeding if given prasugrel: those who had a previous stroke, those who were aged over 75, or those who were underweight. Although this analysis might provide comfort to some physicians, it does not provide insight into what long-term benefit can be achieved by preventing a nonfatal MI.

A recent market survey of 85 interventional and clinical cardiologists reported in Forbes.com surprisingly showed that 36% of their patients would receive prasugrel following angioplasty. Even more striking was the fact that some physicians might use prasugrel for the long-term treatment of ACS patients who did not undergo percutaneous coronary intervention (www.forbes.com/healthcare/2007/11/29/prasugrel-plavix-lilly-biz-healthcare-cx_mh_1129lilly.html

It is clear that patients with suspected ACS with elevated troponin represent a very heterogeneous group. The need to know more about this group of patients is emphasized in the recent consensus document establishing a universal definition of MI, which focuses particularly on the uncertain definition of nonfatal MI in randomized clinical trials. The document emphasizes that we know little about the long term outcome in this mixed patient population. The new definition provides a classification of nonfatal MI for future clinical trials based on the degree of troponin elevation and the mechanism of infarction.

Before we proceed in the direction of prevention of all nonfatal MIs with drugs that have major downside risks, we need to know more about who will benefit. It is imperative that future clinical trials better define nonfatal MI and provide information about the therapeutic benefit of preventing that event.