Current issues of ACP Journal Club are published in Annals of Internal Medicine


Clopidogrel reduced stroke, MI, and vascular death compared with aspirin

ACP J Club. 1997 May-Jun;126:59. doi:10.7326/ACPJC-1997-126-3-059

Source Citation

CAPRIE Steering Committee. A randomised, blinded, trial of clopidogrel versus aspirin in patients at risk of ischaemic events (CAPRIE). Lancet. 1996 Nov 16;348:1329-39.



To compare clopidogrel with aspirin in patients with recent ischemic stroke (IS), recent myocardial infarction (MI), or peripheral arterial disease.


Randomized controlled trial with 1- to 3-year follow-up.


384 clinical centers in 16 countries.


19 185 patients (mean age 63 y, 72% men) who had a recent IS (≤ 6 mo before randomization), a recent MI (≤ 35 d before randomization), or atherosclerotic peripheral arterial disease. Exclusion criteria were age < 21 years, severe cerebral deficit, carotid endarterectomy after qualifying stroke, qualifying stroke caused by carotid endarterectomy, life expectancy < 3 years, uncontrolled hypertension, contraindication to study drugs, potential for pregnancy, or previous participation in other clopidogrel studies. Follow-up was 99.9%.


Patients were allocated to clopidogrel, 75 mg/d (n = 9599), or aspirin, 325 mg/d (n = 9586).

Main outcome measures

The primary outcome was the cluster of IS, MI, or vascular death. Secondary outcomes were amputation in addition to the primary cluster; vascular death; stroke, MI, or all-cause mortality; and all-cause mortality alone.

Main results

Intention-to-treat (ITT) and on-treatment analyses were done. In the ITT analysis, the primary cluster of IS, MI, or vascular death occurred in 939 patients (9.8%) who received clopidogrel and in 1021 patients (10.7%) who received aspirin (P = 0.043). {This absolute risk reduction (ARR) of 0.9% means that 115 patients would need to be treated with clopidogrel (rather than aspirin) for a mean of 2 years to prevent 1 additional occurrence of IS, MI, or vascular death, 95% CI 58 to 8647}*; the relative risk reduction was 8.7%, CI 0.3% to 16.5%. When clopidogrel and aspirin were compared for secondary outcomes, the differences did not reach statistical significance for IS, MI, amputation, or vascular death (10.2% vs 11%, P = 0.076, {CI for the 0.8% ARR -0.1% to 1.6%}*); vascular death (3.6% vs 3.9%, P = 0.29, {CI for the 0.3% ARR -0.2% to 0.8%}*); any stroke, MI, or all-cause mortality (11.8% vs 12.6%, P = 0.081, {CI for the 0.8% ARR -0.1% to 1.7%}*); and all-cause mortality alone (5.8% vs 6%, P = 0.71, {CI for the 0.2% ARR -0.5% to 0.8%}*). Clopidogrel was associated with a greater occurrence of severe rash (P = 0.017) but less gastrointestinal hemorrhage (P = 0.05) than aspirin.


Compared with aspirin, clopidogrel was more effective and as safe in reducing the risk for ischemic stroke, myocardial infarction, or vascular death.

Sources of funding: Sanofi and Bristol-Myers Squibb.

For article reprint: Professor M. Gent, Clinical Trials Methodology Group, Hamilton Civic Hospitals Research Centre, 711 Concession Street, Hamilton, Ontario L8V 1C3, Canada. FAX 905-575-2639.

*Numbers calculated from data in article.


In 1994, a meta-analysis from the Antiplatelet Trialists' Collaboration showed that antiplatelet therapy provided worthwhile protection against serious vascular events (nonfatal MI, nonfatal stroke, or vascular death) in patients at high risk for occlusive arterial disease (1). The most widely tested antiplatelet regimen was medium-dose aspirin (75 to 325 mg/d), but some of the trials assessed other antiplatelet regimens. At the time the overview was done, medium-dose aspirin seemed to be the most effective regimen.

The CAPRIE trial compared a novel antiplatelet drug (clopidogrel) with aspirin. This study of about 20 000 high-risk patients showed that replacing aspirin with clopidogrel was safe and reduced the relative risk for a vascular event by about 9%, which is equivalent to about 5 events/y for every 1000 patients treated. How should these results be interpreted?

The CAPRIE trial certainly establishes clopidogrel as an alternative to aspirin. But the degree to which clopidogrel is actually superior to aspirin remains uncertain because the 95% CI (as discussed in Main results) indicates that the advantages of clopidogrel range from worthwhile to no advantage at all. Because of this uncertainty (and because aspirin is inexpensive and readily available), clopidogrel is likely to be recommended at this time for patients who cannot take aspirin. On the other hand, because aspirin and clopidogrel seem to exert their antiplatelet effects through different mechanisms, combined aspirin and clopidogrel may be better than aspirin alone but without excessive toxicity. If this were to be shown in future randomized trials, then more widespread use of clopidogrel (and aspirin) might be justified.

Colin Baigent, BM, BCh
Radcliffe InfirmaryOxford, England, UK

Colin Baigent, BM, BCh
Radcliffe Infirmary
Oxford, England, UK


1. Collaboration overview of randomised trials of antiplatelet therapy I: Prevention of death, myocardial infarction, and stroke by prolonged antiplatelet therapy in various categories of patients. Antiplatelet Trialists' Collaboration. BMJ. 1994;308:81-106.