Perspective:

One of the strategies suggested for both men and women has been to use aspirin based on 10-year risk for CV events as ≥10%. In this meta-analysis, there was no difference in reductions in the primary CV composite outcome and increases in major bleeding risks based on 10-year CV risk defined as low <10% (median was 6.8%) and high defined as ≥10% (median 12.7%) and diabetes. Remarkably, the authors and JAMA were ready and able to add the three placebo-controlled aspirin studies published as late as November 2018, ASCEND in diabetes, ASPREE in healthy elderly, and ARRIVE in patients with moderate to high risk, none of which established a clear risk/benefit for aspirin. In ARRIVE, the 10-year atherosclerotic CVD risk estimate was 17.3% by the American College of Cardiology/American Heart Association Pooled Estimate, but the observed event rate was <10%. When the analysis was restricted to populations with observation-derived 10-year CV event of ≥10%, aspirin was associated with an absolute risk decrease in CV outcomes of 0.51%, but with an increase of 0.64% for major bleeding. The increase in absolute risk of major GI bleeding and hemorrhagic stroke is similar to the decrease in absolute risk in CV event reduction, each of which is low. While widespread use of aspirin for primary prevention in the middle-aged and elderly needs to be avoided, the decision can be aided by weighing the risk of serious GI bleed (e.g., >75 years, previous history) and intracranial hemorrhage (e.g., hypertension) on an individual basis.