“A daily aspirin taken to ward off heart attacks could do more harm than good,” the Daily Mail warns. It said that aspirin is often prescribed for diabetics as they are at a much higher risk of heart disease. However, a study in 1,276 diabetics found no benefit from either aspirin or antioxodants in preventing heart attacks. It also increases the risk of internal bleeding. BBC News covered the story, and said people who are at high risk and have already had a heart attack or stroke should continue to take it.
This is a reliable study in that it was carefully designed and its outcomes objectively measured. It found that for diabetics, neither aspirin nor the tested antioxidant reduced the risk of having a heart attack, even in the groups usually considered “high-risk”. As reported, there are high-risk groups who will still need the drug, for whom the benefits continue to outweigh the harms. These include patients with diabetes who are known to have heart disease already. The current advice is that anyone taking prescribed aspirin should discuss any concerns they have with their doctor. Local pharmacists should also be able to offer advice.
Where did the story come from?
Professor Jill Belch from the Institute of Cardiovascular Research at the University of Dundee carried out the research with several colleagues who were all part of the Prevention of Progression of Arterial Disease and Diabetes Study Group, Diabetes Registry Group, and Royal College of Physicians Edinburgh. The work was supported by a Medical Research Council grant. The study was published in the peer-reviewed British Medical Journal.
What kind of scientific study was this?
This was a randomised controlled trial conducted in multiple centres. It was designed to test if people with diabetes who take aspirin and antioxidant therapy, together or on their own, are less likely than those on placebo to die from heart attack or stroke, have non-fatal strokes and heart attacks, or have leg amputations due to blockages in the leg.
The researchers recruited 1,276 adult patients with diabetes mellitus from 16 diabetic clinics in Scotland between November 1997 and July 2001. At the beginning of the study, all the participants had evidence of narrowing in the main arteries of the ankle, but had no more major symptoms of blockage. Ultrasound was used to compare the pressure in the arteries at the ankle with the pressure in the arm, so that only those with lower-than-normal ankle pressure were included. The researchers also excluded anyone under 40 years, those who already had symptoms of heart disease or arterial disease, those on aspirin or an antioxidant already, and those with a history of severe indigestion, ulcers, a bleeding disorder or other serious illnesses such as cancer.
This study was double blind and placebo-controlled, which means that neither the researchers nor the participants knew whether they were taking an inactive placebo (control pill) or the active ingredient. The researchers tested two active ingredients, a daily dose of 100mg aspirin as a tablet and an antioxidant capsule containing a mixture of approved antioxidants, including vitamin E, ascorbic acid, pyridoxine, zinc and nicotinamide. Plasma vitamin E and ascorbic acid levels are known to be low in people with diabetes, and both aspirin and the antioxidant therapy are thought to affect platelet stickiness (platelets are blood cells that can clump together and lead to heart attacks).
After participants were randomly allocated to their groups, 320 received the aspirin tablet plus antioxidant capsule; 318 received the aspirin tablet plus placebo capsule; 320 took a placebo tablet plus antioxidant capsule, and 318 took both a placebo tablet and a placebo capsule. This is referred to as a 2x2 design trial.
The researchers measured all the deaths, strokes, heart attacks and amputations, and included them all in a single-outcome measure (the primary event), as well as reporting them separately. Before the study began, the researchers calculated the number of people that would need to take part in order to detect a meaningful difference in the main outcome. They estimated 1,600 participants were needed if they wanted to detect a 25% difference at four years. The risks for each drug were analysed separately, and the researchers looked at how the drugs interacted with each other.
What were the results of the study?
Overall, there was no significant difference between the groups. In the aspirin groups, 116 fatal and non-fatal heart attacks, strokes and amputations (18.2%) occurred compared with 117 (18.3%) in the no aspirin groups.
The 43 deaths from coronary heart disease or stroke occurred statistically at the same rate in the aspirin groups (6.7%) as the 35 deaths in the no aspirin groups (5.5%).
A similar pattern of no statistical effect was seen for the antioxidants. The researchers say they found no evidence for any interaction between aspirin and antioxidant, meaning that the results are unlikely to be caused by one drug interfering with the effects of the other.
What interpretations did the researchers draw from these results?
The researchers conclude that this trial “does not provide evidence to support the use of aspirin or antioxidants” in preventing heart attacks, strokes, amputations or death in the population they studied with diabetes.
What does the NHS Knowledge Service make of this study?
This is a reliable study in that it was carefully designed with randomisation and blinding. All outcomes were also objectively measured. The researchers discuss several aspects to the study:
- Only 1,276 patients were recruited to the trial instead of the 1,600 originally intended. However, the researchers say that they were able to exclude the possibility that the “no difference result” of their trial had occurred by chance. The power calculations (calculations of the numbers of people needed) are an important part of clinical trials such as these
- The 2x2 design, referred to above, allowed the researchers to reduce the number of patients recruited to the trial and still maintain the power to detect a clinically important difference. This was dependent on whether they could prove that the two drugs did not interfere with each other, which they also managed to demonstrate.
- The study did not set out to measure any of the side effects or harms from taking the medications. Although no significant difference was seen in the rate of gastrointestinal bleeding between each group, there was a trend for a greater incidence of gastrointestinal symptoms, including indigestion, in patients given aspirin.
- Between 27% and 33% of participants were smokers, and this may explain why these people were at high risk for the primary events - there was about a 3% risk of an event for each year of the study.
This is an important study as aspirin therapy is prescribed for nearly all patients with type 2 diabetes. This is because they are classified (using standard tools from guidelines) as at high absolute (overall) risk for future heart attack or stroke, and therefore aspirin therapy is usually considered appropriate for most people with the condition. The risk increases in proportion to the number of risk factors that a patient has, such as high blood pressure, high cholesterol, smoking, etc.
A person’s overall risk is thought to be an important determinant of whether they should receive other medications that lower the risk factors, such as statins or blood pressure pills. Aspirin is thought to help at similar levels of risk. This study brings into question current guidance on the use of aspirin in people with diabetes and arterial disease, and in particular the level of risk at which aspirin use is worthwhile. For instance, whether it should only be used by people with known heart disease or if it should also be used by people only thought to be at risk. Further systematic reviews that pool the results of all existing trials by meta-analysis offer hope in answering this remaining question: at what level of risk should people with diabetes be prescribed aspirin?
The current advice is that anyone taking prescribed aspirin should discuss any concerns they have with their doctor. Local pharmacists should also be able to offer advice.
Analysis by Bazian
Edited by NHS Website
Links to the headlines
BBC News, 17 October 2008
The Daily Telegraph, 17 October 2008
Daily Mail, 17 October 2008
The Sun, 17 October 2008
Links to the science
BMJ 2008; Published 16 October