Aspirin prevents cardiovascular events in patients at increased risk of cardiac disease, and aspirin therapy plus clopidogrel (Plavix) adds protection in patients with myocardial infarction and unstable angina and after coronary stenting compared with aspirin alone. However, it is unknown if adding clopidogrel to aspirin prevents atherothrombotic events in high-risk patients. To address this issue, Bhatt and colleagues compared dual anti-platelet treatment with aspirin alone.

CHARISMA (Clopidogrel for High Atherothrombotic Risk and Ischemic Stabilization, Management, and Avoidance), a prospective, randomized, double-blind, placebo-controlled trial, included more than 15,000 patients with clinical cardiovascular disease or multiple cardiac risk factors. Participants received aspirin plus placebo or aspirin plus clopidogrel. Patients were 45 years or older and had one of the following: multiple cardiac risk factors (i.e., two major or three minor or one major and two minor cardiovascular risk factors [see accompanying table on page 1206]), coronary heart disease, cerebrovascular disease, or symptomatic peripheral arterial disease. Patients who met criteria for clopidogrel treatment (e.g., those with a recent myocardial infarction or acute coronary syndrome) were excluded. Patients were receiving previously established treatments (e.g., statins) at the discretion of their physicians.

Patients were followed for a mean of 28 months. The primary end point was a combination of myocardial infarction, stroke, or cardiovascular death, and the primary safety measure was bleeding events. Secondary end points included the primary end point plus hospitalization for unstable angina, revascularization, or transient ischemic attack; individual primary and secondary end points; and all-cause mortality. The average age of participants was 64 years, and most patients were white (80 percent) and men (70 percent). Approximately 75 percent had established cardiovascular disease (i.e., symptomatic), and the remaining participants were included because of risk (i.e., asymptomatic).

There was no significant difference in the primary end point, which occurred in 6.8 percent of the combination group and in 7.3 percent of the aspirin-only group (P = .22). The combined secondary outcome occurred in 16.7 percent of the combination group and in 17.9 percent of the aspirin-only group (P = .04). There was no significant difference in severe bleeding events between the two groups, although moderate bleeding occurred in 2.1 percent of the combination group and in 1.3 percent of the aspirin-only group (P < .001). Subgroup analysis revealed that primary events occurred in 6.6 percent of asymptomatic patients in the combination group and in 5.5 percent of asymptomatic patients in the aspirin-only group (P = .20). This did not occur in symptomatic patients, with 6.9 percent of patients in the combination group and 7.9 percent in the aspirin-only group (P = .046) experiencing primary events. Furthermore, asymptomatic patients in the combination group had an increased risk of all-cause mortality (5.4 versus 3.8 percent, P = .04) and cardiovascular death (3.9 versus 2.2 percent, P = .01)

The authors conclude that clopidogrel added no significant benefit in preventing atherothrombotic events in patients with stable cardiovascular disease or multiple risk factors. The authors estimate a trade-off of 94 prevented ischemic events at a cost of 93 moderate to severe bleeding events. They also conclude that the subgroup observations deserve further study before changes are made to clinical practice. Because of the increased risk of cardiovascular and all-cause mortality in asymptomatic patients, physicians should caution against expanding the use of combined antiplatelet therapy beyond already established indications.

editor’s note: Although clopidogrel has been proven to add benefit in patients with acute events (e.g., myocardial infarction), this benefit did not carry over to patients with stable disease or elevated risk. A related editorial discusses the issue of subgroup findings in the CHARISMA trial.¹ The authors cautioned that, although subgroup analysis suggested benefit or harm, drawing firm conclusions is hazardous. For example, one could suggest a benefit (although small) for the symptomatic subgroup. However, given the major coronary events and significant cardiovascular risk factors that were present in both subgroups, the symptomatic subgroup may not truly be different. Unless future studies indicate otherwise, dual antiplatelet therapy should be avoided for asymptomatic and symptomatic subgroups identified in the CHARISMA trial.—C.C.

1PfefferMAJarchoJAThe charisma of subgroups and the subgroups of CHARISMA [Editorial].N Engl J Med2006;354:1744–6.