NSAIDs should be pharmacy-only: study

woman taking tablet

They increase risk of cardiac arrest and should only be available in pharmacies and in low doses, researchers say

NSAIDs are among the most commonly used drugs worldwide, but their use is associated with a 31% increased risk of cardiac arrest, new research has found.

“Allowing these drugs to be purchased without a prescription, and without any advice or restrictions, sends a message to the public that they must be safe,” says author Professor Gunnar H. Gislason, professor of cardiology at Copenhagen University Hospital Gentofte, Denmark.

“Previous studies have shown that NSAIDs are related to increased cardiovascular risk which is a concern because they are widely used.”

The current study, published in the March issue of European Heart Journal – Cardiovascular Pharmacotherapy, investigated the link between NSAID use and cardiac arrest but did not include non-prescription NSAIDs.

All patients who had an out-of-hospital cardiac arrest in Denmark between 2001 and 2010 were identified from the nationwide Danish Cardiac Arrest Registry.

Data was collected on all redeemed prescriptions for NSAIDs from Danish pharmacies since 1995. These included the non-selective NSAIDs (diclofenac, naproxen, ibuprofen), and COX-2 selective inhibitors (rofecoxib, celecoxib).

A case-time-control design was used to examine the association between NSAID use and cardiac arrest. Each patient served as both case and control in two different time periods, eliminating the confounding effect of chronic comorbidities.

Use of NSAIDs during the 30 days before cardiac arrest (case period) was compared to used of NSAIDs during a preceding 30 day period without cardiac arrest (control period).

Information was not obtained on over-the-counter drugs. Ibuprofen is the only over-the-counter NSAID in Denmark and is limited to small packages of 200 mg dosages.

As patients were their own control, any underestimation of ibuprofen use should be equally distributed between the case and control periods, the researchers say.

A total of 28,947 patients had an out-of-hospital cardiac arrest in Denmark during the ten-year period.

Of these, 3,376 were treated with an NSAID up to 30 days before the event. Ibuprofen and diclofenac were the most commonly used NSAIDs, making up 51% and 22% of total NSAID use, respectively.

Use of any NSAID was associated with a 31% increased risk of cardiac arrest. Diclofenac and ibuprofen were associated with a 50% and 31% increased risk, respectively. Naproxen, celecoxib and rofecoxib were not associated with the occurrence of cardiac arrest, probably due to a low number of events.

“The findings are a stark reminder that NSAIDs are not harmless,” says Professor Gislason.

“Diclofenac and ibuprofen, both commonly used drugs, were associated with significantly increased risk of cardiac arrest. NSAIDs should be used with caution and for a valid indication. They should probably be avoided in patients with cardiovascular disease or many cardiovascular risk factors.”

NSAIDs exert numerous effects on the cardiovascular system which could explain the link with cardiac arrest, he says. These include influencing platelet aggregation and causing blood clots, causing the arteries to constrict, increasing fluid retention, and raising blood pressure.

Professor Gislason says: “I don’t think these drugs should be sold in supermarkets or petrol stations where there is no professional advice on how to use them. Over-the-counter NSAIDs should only be available at pharmacies, in limited quantities, and in low doses.”

“Do not take more than 1200 mg of ibuprofen per day,” he continued. “Naproxen is probably the safest NSAID and we can take up to 500 mg a day.

“Diclofenac is the riskiest NSAID and should be avoided by patients with cardiovascular disease and the general population. Safer drugs are available that have similar painkilling effects so there is no reason to use diclofenac.

“The current message being sent to the public about NSAIDs is wrong.

“If you can buy these drugs in a convenience store then you probably think ‘they must be safe for me’.

“Our study adds to the evidence about the adverse cardiovascular effects of NSAIDs and confirms that they should be taken seriously, and used only after consulting a healthcare professional.”

Previous Women at higher risk of chronic lung disease
Next Guild calls on Hunt to deliver

NOTICE: It can sometimes take awhile for comment submissions to go through, please be patient.


  1. Big Pharma

    How pharmacy professional bodies were unable to prevent the previous down scheduling of NSAIDs is beyond me. Evidence is clear. In the HMR setting I often see duplicated and sometimes triplicated NSAID therapy. A number of these patients over the years have been recently discharged from hospital with GI bleeds or cardiovascular complications. BUT HEY…who needs HMRs right PGA?

    • Willy the chemist

      …who needs HMRs right PGA
      That’s not really a fair comment. As I remembered, there were a number of HMR operators gaming the system. It was the same with MedChecks, with a few members of the premier discount group running them into strastopheric numbers.

      • Big Pharma

        90 pharmacists across the entire country were claiming 20 HMRs/month or more (that’s part-time at best). Would have cost nothing to audit 90 people. Even quicker (fewer) once you assess who is doing >20 HMRs/week. Yet to see the break down of these claims as the PGA will not release this information. They just sat on the information until program funds were nearing exhaustion in an attempt to manipulate funding. Obviously, funding should be separate from CPA.
        Yes indeed medschecks was the biggest rort with some stores doing 2000/week! Still awaiting follow-up….unpunished it would seen. I would be interested in seeing how many outcomes (if any) or changes to therapy resulted from these services. Worth noting a referral is needed for a HMR and an accountable report is written.
        Some pharmacies game the PBS. Suppling all repeats of all allowable medications for all SN patients before the year expires. Some pharmacies open non-PBS stores in competitive areas and fill the PBS paperwork off site in a different location (with a provider number). Many pharmacies promote non-evidence based medications and push companion sales. Should we maybe cap the number of PBS items that can be dispensed per day? Maybe remove all complementary medications void of evidence from the shelves? That is the brainwave behind the HMR cap.

    • Tim Hewitt

      But HEY! PGA is 100% responsible for introducing HMR.. ! get a name ‘Big Pharma’!.. too gutless to use you own name??

      • Big Pharma

        Evidence of cost-saving is responsible for the introduction of HMRs, the PGA merely supported the evidence. In fact evidence (and community outrage) was the only reason the PGA was unable to force a complete program moratorium. Interesting how quickly their position swung when the middle-man (their members) weren’t entitled to payment for simply “passing on the referral”. Not surprisingly focus quickly switched to non-evidence based “medschecks”.
        Direct referrals only came about due to dangerous delays in patient review times as pharmacies tendered out referrals for maximise profit. In fact, under the old referral pathway, I previously received referrals that were 6 months old….not easy trying to explain to a family that you have just received a referral for someone that passed away 4 months earlier.

  2. David Haworth

    Be nice to read the published study AJP. Hows about a link if possible. %50 increase risk for diclofenac users. That could be 1 in 10,000 increasing to 2 in 10,000. Whats the real risk?

    • Willy the chemist

      Quoting % increase in risk is an example of headline bias. In my opinion, reporters have to be more accountable. Risk should be quoted in both relative (%) and absolute (1 in 20,000).
      What is 31% increase risk of cardiac arrest? Really 31%? Absolute 31% increase with no margin of error?
      Researchers/academicians present their findings as absolute facts when sometimes these can be open to interpretations as well.

  3. Drugby

    This is an earlier study by the same author, showing most NSAIDs are associated with an increased cardiovascular risk, both in healthy people and among patients with known cardiovascular disease: http://onlinelibrary.wiley.com/doi/10.1111/bcpt.12244/epdf

    This data just reinforces that NSAIDs should be used in the lowest dose for the shortest duration if clinically indicated. And that pharmacists can help determine the least risky NSAID for an individual based on their medical history and cardiac/GI risk.

  4. Drugby
  5. Tony Pal

    To get some perspective on this issue visit the TGA DAEN site


    type in ibuprofen, select ALL for products and select ~10 years eg Dec 2006 to Dec 2016, you will get 18 pages of report which you can print and save as pdf, which is searchable eg ‘cardiac’, ‘myocardial infarction’, ‘stroke’. Cardiac issues do not appear until page 8, with no deaths. Check out Column 5 for deaths and see what is really causing NSAID related deaths. namely dependence, drug abuse, and overdose, all on Page 1. Perform the same task for Diclofenac.
    That cardiac issues can occur with NSAIDs has been well-known since 2004. The recent news is more to do with the significance of cardiac event findings 31% & 51 % and that should be related to the 100% increased risk (1.96 Xs placebo) for rofecoxib, where the latter was removed from the market. In my personal opinion, the news is nothing new, there is a bit of overhyping of the significance of cardiac events in relation to more important issues that would justify returning the product to Pharmacy only. However, given the number of deaths over 10 years, you could equally argue that the short term safety is well proven, for the general sale of small packs not intended for long term use.

Leave a reply