Heart and lungs

Heart risk of painkillers examined

A painkiller taken by millions can increase the risk of heart attack and stroke by 40%, the Daily Mail has today reported. The newspaper says that researchers are calling for the drug, called diclofenac, to be available on prescription only.

The news is based on a large review that looked at the cardiovascular risks associated with a class of widely used painkillers called non-steroidal anti-inflammatory drugs (NSAIDs). NSAIDs in high-dose formulations are usually only available on prescription, but some low-dose NSAIDs, including ibuprofen, naproxen and diclofenac, can be bought over the counter.

The review found that diclofenac raised the risk of heart problems by 22% when taken at over-the-counter doses and by 40% at prescription strength. Naproxen and low-dose ibuprofen were least likely to increase the risk of heart attacks and strokes.

While previous research has highlighted the cardiovascular risks of some NSAIDs, this review of observational studies provides some important new information about the risks associated with all currently available NSAIDs at different doses. As such, its findings will no doubt be important to future decisions about how these drugs should be used and regulated.

However, it is important to note that for a healthy individual who takes diclofenac, the increased risk to the heart is still very small. The nature of this research means it is not possible to estimate accurately how small this risk is. Anyone who is concerned about taking NSAIDs should not stop taking these drugs but should consult their doctor.

Where did the story come from?

The study was carried out by researchers affiliated with Hull York Medical School, the Institute for Clinical Evaluative Sciences, the University of Toronto in Canada and the University of Newcastle in Australia. It received no external funding. The study was published in the peer-reviewed journal PLoS Medicine.

The research was covered fairly in most newspapers. In its print version of the story the Daily Mail featured a large front-page headline warning of a “Painkiller heart alert”, which may have been alarming. However, within the article itself the Daily Mail did feature prominent messages that patients should not panic and should not stop taking their medication. Both the Daily Mail and The Daily Telegraph reported that, for most healthy people, the increased risk of heart and other problems from diclofenac was small, and the reports featured in the Daily Mail, The Daily Telegraph and the Daily Express all included comments and advice from independent experts.

What kind of research was this?

This was a systematic review comparing the risks of individual NSAIDs taken at typical doses by people at home, rather than in hospital. The researchers say that there are concerns about the risk associated with non-prescription NSAIDs available in low-dose forms, such as ibuprofen, naproxen and diclofenac.

The researchers point out that while some randomised trials have highlighted the cardiovascular risk of some NSAIDs, little is known about how the risks of individual drugs compare when used at different doses, for different lengths of time and in different populations. For this reason the researchers set out to examine the outcomes seen in controlled observational studies, which would better reflect the risks associated with the typical domestic use of NSAIDs rather than the risks associated with their use in the idealised setting of a clinical trial. To date, randomised trials of NSAIDs have reported only small numbers of heart and stroke problems.

What did the research involve?

The researchers searched a wide range of electronic databases for relevant studies published between 1985 and 2010 that had reported on the cardiovascular risks associated with the use of individual NSAIDs in population settings. They included only non-randomised, controlled observational studies in their literature search. These observational studies included case control, cohorts and case-crossover studies. They then assessed the methodological quality of the selected studies. From a total 459 potentially relevant papers, 51 studies met their criteria.

From the studies gathered, the researchers extracted and pooled information about the risk of major cardiovascular events associated with individual NSAIDs. They also assessed subsets of studies that provided relevant information to examine the risk of NSAIDs in different doses and in people with low and high existing risk of heart problems. To compare different drugs they carried out a further type of analysis, called a pair-wise comparison, where they indirectly compared each drug against another in turn, taking the results from separate trials.

The overall analyses included data from 30 case-control studies and 21 cohort studies involving more than 2.7 million individuals and featuring a total of 184,946 cardiovascular events.

What were the basic results?

The researchers looked at the drugs where there were 10 or more studies. Of drugs where there were 10 or more studies, researchers found that the highest overall risks were seen with rofecoxib and diclofenac, and the lowest with ibuprofen and naproxen. Compared with not using any NSAIDs, the researchers found:

  • rofecoxib increased the risk of heart problems by 45% (95% CI 1.33 to 1.59)
  • diclofenac increased the risk by 40% (95% CI 1.27 to 1.55)
  • ibuprofen increased the risk by 18% (95% CI 1.11 to 1.25)

In a subset of studies that looked at risk associated with lower doses they found:

  • low doses of rofecoxib increased the risk by 37% (95% CI 1.20 to 1.57)
  • low doses of celecoxib increased the risk by 26% (95% CI 1.09 to 1.47)
  • low doses of diclofenac increased the risk by 22% (95% CI 1.12 to 1.33)

It is important to note that the drug rofecoxib has already been withdrawn from the market because of its association with a raised risk of cardiovascular events. Including it in the study allows the risk associated with other drugs to be compared with the risks of rofecoxib.

Ibuprofen only posed a risk when taken at a higher dose and naproxen had no significant risk at any dose.

The researchers say the increase in risk was proportional for both high- and low-risk groups. This means that, relative to their risk if not using NSAIDS, the risks for both groups increased to the same extent. The risk of cardiovascular problems also rose early in the course of treatment. For some NSAIDs, risk was found to increase within the first month of taking the drug.

How did the researchers interpret the results?

The researchers say the results of their review “are robust enough to inform clinical and regulatory decisions”.

  • They call for “regulatory action” on diclofenac, as it is currently available without prescription.
  • They say the limited data on etoricoxib “raises serious concerns” about safety, particularly as similar drugs such as rofecoxib have been withdrawn.
  • They say that, in the case of ibuprofen, labelling warnings should be strengthened to stop patients who are already at high risk of cardiovascular problems from exceeding the maximum recommended dose.
  • They question the continued use of indomethacin.


This large review has published some important information on the cardiovascular risks associated with NSAIDs, including the risk associated with different doses and in populations at both high and low risk of cardiovascular events. It raises concerns about some of these risks, in particular the risk associated with the widely used non-prescription drug diclofenac.

As its authors point out, it had some limitations.

  • It had to rely on observational studies (rather than randomised controlled trials), which are subject to bias, especially in terms of other factors (confounders) that might influence results. However, the researchers did take steps to minimise this risk.
  • The data in the studies mainly came from large administrative databases and electronic health records, and may not have been comprehensive, especially concerning key information such as use of non-prescription NSAIDs and aspirin, or information about people’s risk of heart problems.
  • The review suffered from ‘heterogeneity’. This means that many of the studies varied in their design, their methods and how they analysed results. Heterogeneity makes it harder to combine the results of different studies accurately and can, therefore, throw doubt on the findings of systematic reviews.

Patients using NSAIDs who are worried about side effects should not stop taking them, but instead consult their doctor.

NHS Attribution