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


Low-dose aspirin did not alter the incidence or delay the onset of angina pectoris when used as primary prevention in men

ACP J Club. 1991 Mar-April;114:39. doi:10.7326/ACPJC-1991-114-2-039

Source Citation

Manson JE, Grobbee DE, Stampfer MJ, et al. Aspirin in the primary prevention of angina pectoris in a randomized trial of United States physicians. Am J Med. 1990 Dec;89:772-6.



To determine the effect of low-dose aspirin (325 mg on alternate days) on the primary prevention of angina pectoris.


Randomized, double-blind, placebo-controlled trial with participants followed for a mean of 60.2 months (range 45.8 to 77.0 mo).


The Physicians' Health Study, a study designed to evaluate the effects of aspirin on reduction of risk for cardiovascular disease and of β-carotene in the prevention of cancer. Data were collected through the use of mailed questionnaires, with follow-up data on mortality and morbidity 100% and 99.7% complete, respectively.


21 738 male physicians aged 40 to 84 years, free from angina pectoris at baseline, with no previous myocardial infarction, stroke, or transient cerebral ischemia.


Aspirin, 325 mg (n = 10 859), or placebo (n = 10 879) was given on alternate days.

Main outcome measure

A report of angina pectoris confirmed by a report of a positive diagnostic test (confirmatory exercise tolerance test, exercise thallium, exercise radionuclide ventriculogram, or narrowing ≥ 50% of a coronary artery on arteriogram), or coronary revascularization procedure.

Main results

331 participants had confirmed angina pectoris, including 194 patients who had a coronary revascularization procedure. 173 patients (1.6%) in the aspirin group had confirmed angina pectoris compared with 158 patients (1.4%) in the placebo group { P = 0.4; absolute risk difference 0.2%, 95% CI -0.2% to 0.5%}.* 105 patients (1.0%) in the aspirin group compared with 89 (0.8%) in the placebo group had coronary revascularization { P = 0.2; absolute risk difference 0.2%, CI -0.1% to 0.4%}.* When controlling for coronary risk factors, the relative risks were 1.07 (CI 0.84 to 1.36) and 1.11 (CI 0.81 to 1.52) for angina pectoris and coronary revascularization, respectively. Relative risk did not change substantially between the first and fifth year of treatment.


Low-dose aspirin did not alter the incidence or delay the onset of confirmed angina pectoris.

Source of funding: National Institutes of Health.

Address for article reprint: Dr. J.E. Manson, 55 Pond Avenue, Brookline, MA 02146, USA.

*Numbers calculated from data in article.


In previous reports from the Physicians' Health Study (PHS), the investigators reported that low-dose aspirin reduced the risk for a first myocardial infarction and the occurrence of self-reported migraine. The current report shares problems with the earlier analyses, while introducing some new ones. Cardiovascular end points occurred at just 15% of the rate anticipated by the investigators, a result of the healthy lifestyles of U.S. physicians and of the study exclusion criteria. In addition, patients were not accepted for the study unless they passed a compliance "test." This further restricts the generalizability of the results: High adherers may have more favorable study outcomes regardless of treatment (1).

One should also note 2 considerations that are distinctive to this report. Angina pectoris was counted only if a positive diagnostic test was done or if a coronary revascularization procedure was done. This restricted end points to the severe end of the spectrum of all patients with angina pectoris, because patients with milder symptoms are less likely to be tested. A second issue concerns the authors' suggestion that "...the role of platelets in the occlusion of coronary arteries leading to infarction may be different from their role in the development of atherosclerotic stenosis." This conclusion may have been predetermined by their decision to exclude from analysis all participants who reported a first myocardial infarction. If angina and myocardial infarction were treated as a combined end point, aspirin would appear beneficial, and the role of platelets may have been viewed as similar for infarction and atherosclerotic stenosis.

Ralph I. Horwitz, MD
Yale University School of MedicineNew Haven, Connecticut, USA


1. Horwitz RI, Viscoli CM, Berkman L, et al. Treatment adherence and the risk of death after a myocardial infarction. Lancet. 1990;336:542-5.

Authors' Response

The commentator raises several issues related to the generalizability of study results in the Physicians' Health Study (PHS). It is important to emphasize that the issue of generalizability is distinct from, and secondary to, the issue of internal validity of a study. The specific design features of the PHS, including the eligibility criteria and the "compliance test," actually serve to enhance the internal validity of the trial. In the present analyses, our goal was to examine the effect of chronic platelet inhibition on atherosclerotic progression, as distinct from its effect on thrombotic occlusion. Therefore, angina pectoris was examined separately from myocardial infarction. A combined end point would not have permitted the pathogenetic distinction we sought.*

JoAnn E. Manson, MD
Paul M. Ridker, MDCharles H. Hennekens, MD

*An interesting point...but if atherosclerotic progression and thrombotic occlusion are part of the same process, then Dr. Horwitz is correct — The Editor