Heart and lungs

Painkillers and risk of heart problems

“Taking painkillers increases risk of death to heart attack victims by 55%,” reported the Daily Mail. The newspaper quoted an author of new drug research who said their results indicated that there is ‘no apparent safe therapeutic window’ for patients with prior heart attack to take NSAID painkillers, a class of drugs that includes ibuprofen.

The research used information gathered on 100,000 Danish people who had experienced their first heart attack between 1997 and 2006, calculating whether their use of NSAIDs (non-steroidal anti-inflammatory drugs) related to their risk of death or a second heart attack. The study found even short-term use was associated with an increased risk compared to not using the drugs, although the study could not calculate factors such as how dosage related to risk.

Current UK guidelines already state that NSAID drugs must only be used cautiously in people with heart conditions and should not be used at all in some cases. Previous research has recognised that NSAIDs increase the risk of cardiovascular complications including heart attacks and strokes. This is in line with this important study’s findings that users of the drug had a higher risk of death or recurrent heart attack than non-users.

People with a history of cardiovascular problems such as stroke and heart attack should consult their doctor if they need to take painkillers, as they can advise them on their appropriate options.

Where did the story come from?

The study was carried out by researchers from Copenhagen University Hospital. No sources of funding were reported. The study was published in the peer-reviewed medical journal Circulation.

The study was covered accurately by The Daily Telegraph and Daily Mail . It should be highlighted that the 55% risk increase figure reported in the Daily Mail refers to the risk of death or recurrent heart attack rather than death alone. The increased risk of death with up to 90 days of NSAIDs use was 56%. Although the newspapers reported the relative risks between NSAID users and non-users, neither newspaper stated the absolute risk of a heart attack in the study; that is, how common heart attacks were in the study population as whole.

What kind of research was this?

This was a prospective cohort study that looked at whether taking non-steroidal anti-inflammatory drugs (NSAIDs) after a heart attack was associated with a greater risk of a heart attack and death in the subsequent period.

NSAIDs (such as ibuprofen) are a commonly used and safe type of painkiller normally used for short periods. The use of NSAIDs is discouraged in patients with established cardiovascular disease such heart failure or who have had a heart attack. However, if their use is unavoidable it is advised that people use them for as short a time as possible.

This study wanted to examine the risks of using NSAIDs after heart attack to see if there was a safe period that people could take the painkillers following their heart attack.

What did the research involve?

This study was carried out in Denmark, where each resident has a unique identification number that allows their medical records to be linked across directories. From hospital admission directories the researchers found people who had been admitted for a heart attack for the first time between 1997 and 2006, and had survived it. They then linked the participants’ profiles to a drug prescription database in order to see which prescribed medications each person had claimed, including NSAID drugs.

NSAIDs are a broad class of drugs that include non-selective NSAIDs, such as ibuprofen, diclofenac and naproxen, and ‘selective inhibitors of cyclo-oxygenase-2’ (COX-2 inhibitors) that include rofecoxib and celecoxib. All COX-2 inhibitors are, in fact, completely contraindicated (deemed unsuitable for use) in people with coronary heart disease.

The researchers also looked at medications that people with a heart condition were likely to be prescribed, such as beta-blockers, ACE inhibitors, diuretics and anti-diabetic drugs. They worked out the dose and treatment duration by looking at how often the drugs was prescribed and how many were prescribed each time. They also calculated an estimate of the average number of pills taken each day for each prescription period.

High doses of each drug were defined as being above the recommended minimal dose for each drug. For each NSAID these high doses were considered to be:

  • ibuprofen: over 1200mg
  • diclofenac: over 100mg
  • naproxen: over 500mg
  • rofecoxib: over 25mg
  • celecoxib: over 200mg

The researchers used data on comorbidities (other health conditions the participant had) in order to determine how much these they would affect the outcome seen over the follow-up period. Additionally they used data on each individual’s salary to estimate their socioeconomic status.

What were the basic results?

The researchers found that a total of 102,138 patients had been admitted with a first-time heart attack between 1997 and 2006. Of these 83, 675 had survived and were included in the study. The participants were on average 68 and 63% were men.

Forty-two percent of people had claimed at least one prescription for any type of NSAID after their discharge.

In the follow-up period there were 35,257 subsequent heart attacks (fatal and non-fatal) and 29,234 deaths. The study did not report what the average follow-up was for each person, instead it presented the number or deaths or second heart attacks per 1,000 person years. This means they added up the total number of follow-up years each participant was followed. For example, a study following 300 people for 10 years would generate 3,000 person years of data. From this data the researchers calculated the risk of death or second heart attack in people who had claimed a prescription for a NSAID compared to those who had not.

They found that taking an NSAID of any type for up to a week was associated with a 45% increased risk of death or recurrent heart attack relative to not taking these drugs (hazard ratio [HR] 1.45, 95% confidence interval [CI] 1.29 to 1.62). People who received NSAIDs for longer periods also had an increased risk: use for 90 days was associated with 55% increased risk (HR 1.55, 95% CI 1.46 to 1.64).

They then analysed the effect of taking individual NSAIDs for up to seven days, comparing the risk with taking no prescription. Diclofenac and naproxen were found to increase the risk of death or heart attack:

  • diclofenac: increased risk threefold  (HR 3.26, 95% CI 2.75 to 3.86)
  • naproxen: increased risk by 76% (HR 1.76, 95% 1.04 to 2.98)

There was no increased risk associated with up to seven days’ of treatment with rofecoxib, celecoxib or ibuprofen.

However, using rofecoxib, celecoxib and ibuprofen for 7-14 days was associated with an increased risk relative to people who had not received a prescription:

  • rofecoxib: two-fold increased risk  (HR 2.27, 95% CI 1.69 to 3.04)
  • celecoxib: 90% increased risk (HR 1.90, 95% CI 1.46 to 2.48)
  • ibuprofen: 50% increased risk (HR 1.50, 95% CI 1.24 to 1.82)

How did the researchers interpret the results?

The researchers said that in patients with prior heart attack ‘short term treatment with most NSAIDs is associated with increased cardiovascular risk’. They highlighted that diclofenac was particularly associated with an increased risk and is available over-the-counter in some countries. They say that there is ‘no apparent safe therapeutic window for NSAIDs in patients with prior heart attack’. They add that their study challenges the ‘current recommendations of low-dose and short-term use of NSAIDs being safe’.


This was a large cohort study that used data from a number of Danish health registries. It found that in this population it was not safe to take NSAIDs for a short period after heart attack.

It is worth considering when interpreting these results, that in the UK, prescribing advice already recommends cautious use of NSAIDs in people with heart conditions. The advice is that the lowest effective dose should be prescribed for the shortest period needed to control symptoms and the need for long-term treatment should be reviewed periodically. In people with severe heart failure, guidelines state they should not be used at all. Also, all COX-2 inhibitor drugs are completely contraindicated in people with coronary heart disease or any other cardiovascular condition.

The participants in this study experienced their first heart attack between 1997 and 2006, and it is possible they would have been prescribed higher NSAID doses than people currently are, given today’s greater knowledge of the contraindications of these drugs in this population. Also, the study did not analyse how the size of the dose related to risk, and was carried out in the Denmark, where prescribing practices and doses may differ from the UK.

The researchers also highlighted additional limitations of the study design.

  • The study did not collect information about important clinical parameters such as blood pressure, BMI, smoking habits and lung function, so there may be some unmeasured confounders that contribute to the effect.
  • They suggest that patients prescribed NSAIDs may have been done so because of underlying health problems that were less common in those not treated with the drugs. While adjustments were made to account for this confounding factor, they may not have been adequate.
  • The researchers estimated the dose and duration of NSAID use by looking at prescription claims information. A potential problem is that a filled prescription does not tell us how or whether the medication was taken. People also do not necessarily take their medication consecutively (for example, they could have spread out the prescribed dose over a much longer period than indicated) or follow the attached guidance. Furthermore, the study is unlikely to capture over-the-counter NSAIDs taken without prescription.
  • The study did not break down the analyses by dose, meaning it is not possible to quantify the risk with any particular dose of NSAID.

This study supports the existing advice that NSAIDs be used with caution in people with cardiovascular conditions. People who have experienced prior heart attack or cardiovascular problems should consult their GP about the appropriate painkiller to use if pain relief is required for any reason.

NHS Attribution