Intended for healthcare professionals


NSAIDs and the failing heart

BMJ 2016; 354 doi: (Published 28 September 2016) Cite this as: BMJ 2016;354:i5163

Chinese translation


  1. Gunnar H Gislason, professor of cardiology1 2 3,
  2. Christian Torp-Pedersen, professor of cardiology4
  1. 1Cardiovascular Research Centre, Department of Cardiology, Copenhagen University Hospital Herlev and Gentofte, 2900 Hellerup, Denmark
  2. 2National Institute of Public Health, University of Southern Denmark, Copenhagen, Denmark
  3. 3Danish Heart Foundation, Copenhagen, Denmark
  4. 4Department of Health Science and Technology, Aalborg University, Aalborg, Denmark
  1. Correspondence to: G H Gislason gg{at}

NSAIDs pose a clear risk to some patients and tighter regulation is justified

Non-steroidal anti-inflammatory drugs (NSAIDs) are among the most commonly used drugs and are mainly used to alleviate pain and inflammation. NSAIDs were first introduced over a century ago and most of the commonly used NSAIDs were registered in an era with few requirements for safety documentation. Therefore, for most of these drugs, there is an alarming lack of randomised studies addressing their safety. With the introduction of selective cyclo-oxygenase (COX) 2 inhibitors, concerns were raised that unbalanced COX 2 inhibition might increase cardiovascular risk.1 This has been demonstrated in several randomised and observational studies, and further warning flags have been raised about the cardiovascular safety of non-selective NSAIDs.2 3 4 5 6 7

Arfé and colleagues8 present results from a nested case-control study based on real world data from four European countries and almost 10 million NSAID users. The researchers found that NSAIDs raise risk of hospital admission for heart failure and that the magnitude of risk varied between individual NSAIDs and according to the dose prescribed. A cause of concern was that commonly used NSAIDS such as ibuprofen, diclofenac, and naproxen were all associated with increased risk of heart failure and that use of high doses of ibuprofen and diclofenac doubled the odds, while the increased risk associated with use of high dose naproxen was slightly lower. By contrast with other selective COX 2 inhibitors (eg, rofecoxib, etirocoxib), celecoxib was not associated with increased risk of heart failure. But celecoxib was used mostly in low doses, so the safety of higher doses of celecoxib remains unexplored.

The study adds to the bulk of evidence linking NSAID use with increased risk of heart failure, and further provides novel evidence on a dose-response relation.7 9 The main strength of the study was the study size, linking information from four different European countries, and the consistency across countries. Although the researchers found that the risk of heart failure associated with NSAID use was independent of history of heart failure, other studies have demonstrated a particularly high cardiovascular risk associated with NSAID use in people with heart failure.10 Therefore, a continued restrictive approach towards use of NSAIDs in this group is recommended.

However, the clinical perspective of the study was limited: it only reported odds ratios and did not provide data on excess absolute risk. Information on relative risk or odds ratios is not necessarily meaningful without knowing absolute risk, and hard to communicate to the public or patients needing NSAIDs. Notably, the magnitude of excess absolute risk depends on the baseline cardiovascular risk; therefore, a larger effect is expected in high risk populations, such as with established heart failure or a high cardiovascular risk profile.

Information on absolute risks is valuable for clinicians and patients evaluating the balance between benefit and harm of treatment. Low risk patients might accept the small additional risk associated with treatment while higher risk patients might prefer to consider alternative treatments. The value of physiotherapy and exercise on arthritic pain is well known and in some patients other pain treatments, such as paracetamol or a weak opiate, might be a good choice. For patients who do need NSAID treatment, it is important to consider the different risk profiles of the individual drugs. The selective COX 2 inhibitors and diclofenac have repeatedly been associated with higher cardiovascular risk, and therefore it seems prudent to avoid them and consider lower risk naproxen at the lowest effective dose.

Owing to the widespread use of NSAIDs, even a small increase in cardiovascular risk is a concern for public health. The European Medicines Agency and the US Food and Drug Administration have issued warnings about the potential harmful effect of NSAIDs, and in particular diclofenac.11 12 13 The same concerns have been raised in a recent position paper by the European Society of Cardiology, which recommends against any use of diclofenac.14 However, NSAIDs are still widely available over the counter in supermarkets and convenience stores without any provision of professional advice on their use or potential adverse effects. This practice further fuels the common misconception that NSAIDs are harmless drugs that are safe for everyone. Therefore, a more restricted policy by regulatory authorities on the availability of NSAIDs and requirements for healthcare professionals providing advice on their use and potential harm is warranted.


  • Research, doi: 10.1136/bmj.i4857
  • We have read and understood the BMJ policy on declaration of interests and declare the following interests: CT-P advises Bayer on anticoagulation for atrial fibrillation.

  • Provenance and peer review: Commissioned, not externally peer reviewed.


View Abstract