Benefit of aspirin for healthy people is uncertain

Researchers at the Clinical Trial Service Unit at the University of Oxford analysed data from a number of primary and secondary prevention trials that had compared long-term aspirin use against controls. The findings are published in The Lancet.

In the primary prevention trials, aspirin reduced the risk of a non-fatal heart attack by about one fifth. This corresponds to five fewer such episodes each year for every 10,000 people treated. This is offset by a comparable increase in bleeds with long-term aspirin use. One extra stroke is caused by bleeding and three extra gastrointestinal bleeds occur each year per 10,000 treated.

In the secondary prevention studies, aspirin reduced the risk of a serious vascular event (a heart attack, stroke or cardiovascular death) by about a fifth. But the risk of an event is much higher among people with established cardiovascular disease, so that there were 150 fewer such events each year for every 10,000 patients treated. This large benefit greatly exceeds the risk of bleeding. 

In both sets of trials, the risk of a serious vascular event was reduced to a similar degree in both men and women.

Previous reviews of primary prevention trials have led to guidelines recommending that aspirin be used widely among healthy people who are more at risk of coronary heart disease, having raised blood cholesterol or blood pressure for example.

But the new analyses show that many people with above average risk of coronary heart disease are also at above average risk of suffering a bleed, so this method of selecting whom to treat may not be appropriate. 

Professor Colin Baigent, an MRC scientist who led the work at the Clinical Trial Service Unit, says: ‘The primary prevention trials were completed some years ago, when modern drugs such as statins were not widely available. Nowadays, primary prevention with statins and other drugs can safely half the risk of heart attacks and strokes.’

‘When aspirin is added to such drugs, the further reduction in serious vascular events is only about half as large as when it is used alone, but the bleeding risks will remain about the same. This has important implications when judging the likely effects of aspirin in practice.’

The authors conclude: ‘Aspirin is of clear benefit for people who already have cardiovascular disease, but the latest research does not seem to justify general guidelines advocating the routine use of aspirin in all healthy individuals above a moderate level of risk for coronary heart disease.’

When prescribing aspirin to healthy individuals, it is important to consider the potential of such a policy to cause harm. Professor Baigent adds: ‘Drug safety really matters when making recommendations for tens of millions of healthy people. We don’t have good evidence that, for healthy people, the benefits of long-term aspirin exceed the risks by an appropriate margin.’