NSAIDs and risk of acute myocardial infarction

2 downloads 0 Views 105KB Size Report
in cardiovascular risk between etoricoxib and diclofenac.4 The magnitude of cardiovascular risk, for other nonselective NSAIDs with avail- able data, lies ...
PRACTICE POINT www.nature.com/clinicalpractice/rheum

NSAIDs and risk of acute myocardial infarction Original article Singh G et al. (2006) Risk of acute myocardial

infarction with nonselective non-steroidal anti-inflammatory drugs: a meta-analysis. Arthritis Res Ther 8: R153

SYNOPSIS KEYWORDS cardiovascular risk, NSAIDs

on the generic inverse variance method) for nonselective NSAIDs as a drug class was used for the analysis, and meta-analyses on individual NSAIDs were also carried out where data were available. A pooled measure of acute myocardial infarction risk was calculated from the individual studies’ estimates of acute myocardial infarction risk relative to past NSAID use and no NSAID treatment.

D P E C T N EC R R F O O C O N R U P BACKGROUND

Traditional nonselective NSAIDs are commonly prescribed to relieve symptoms of pain and inflammation, and have been favored over cyclo-oxygenase (COX)2 inhibitors because of the increased risk of cardiovascular adverse events associated with the latter. Recent studies have indicated a possible association between nonselective NSAIDs and an increased risk of cardiovascular thrombotic adverse events (e.g. acute myocardial infarction), which is of great concern. OBJECTIVE

The objective of this study was to determine the relative risk of acute myocardial infarction in patients who receive treatment with traditional, nonselective NSAIDs, compared with patients with past exposure to NSAIDs or no treatment with NSAIDs. DESIGN AND INTERVENTION

This meta-analysis included data from studies published from January 1980 to June 2005. Candidate studies were identified in all major electronic databases such as MEDLINE, BIDS and EMBASE, and studies included in the analysis had to contain data from population databases that compared risks of acute myocardial infarction in patients treated with traditional, nonselective NSAIDs, patients with past exposure to NSAIDs, or no NSAID treatment. A random-effects model (based

2 NATURE CLINICAL PRACTICE RHEUMATOLOGY

OUTCOME MEASURES

The primary outcome of this analysis was the clinically confirmed prevalence of acute myocardial infarction in the analyzed studies. RESULTS

A total of 14 studies were included in this analysis. Overall, nonselective NSAIDS were associated with an increased risk of acute myocardial infarction (relative risk 1.19, 95% CI 1.08–1.31) compared to patients with past use of NSAIDs or no previous NSAID use. Similar results were observed in the analysis of individual nonselective NSAIDs: for diclofenac the relative risk was 1.38 (95% CI 1.22–1.57), and for ibuprofen the relative risk was 1.11 (95% CI 1.06–1.17). Naproxen, however, was not associated with increased risk of acute myocardial infarction (relative risk 0.99, 95% CI 0.88–1.11). CONCLUSION

The authors conclude that there is evidence to suggest that some nonselective NSAIDs are associated with an increased risk of acute myocardial infarction. Analyses of the risks associated with individual NSAIDs indicated that risk is increased by treatment with diclofenac and ibuprofen, but not with naproxen.

XXX 2007 VOL 3 NO X

PRACTICE POINT www.nature.com/clinicalpractice/rheum

C O M M E N TA RY Luis A García Rodríguez* and Antonio González-Pérez The overall safety profile of NSAIDs has received much attention. Gastrointestinal complications that are associated with this class of drugs have been known for a long time. Selective COX2 inhibitors, the introduction of which was expected to help avoid the gastrointestinal complications associated with NSAIDs, have now raised new concerns on cardiovascular safety. It has since been discovered that traditional (i.e. nonselective) NSAIDs could also share some of the cardiovascular risks associated with selective COX2 inhibitors. The body of evidence in this regard consists of data from clinical trials and from an increasing number of observational studies, most of which have used patient data from large databases such as the General Practice Research Database in the UK, or administrative databases in the US or Canada. At first glance, clinical trials, by virtue of their randomized treatment allocation, provide firmer evidence than observational studies do. The type of individual NSAIDs, the prescribed dose and the duration of response, however, are often limited in randomized clinical trials (RCTs). Kearney and colleagues summarized evidence from RCTs that compared selective COX2 inhibitors with placebo or nonselective NSAIDs.1 Overall, COX2 inhibitors were associated with a 40% increased risk of cardiovascular events compared with placebo, and with a more than 50% increased risk compared with naproxen. No difference in risk was found, however, when COX2 inhibitors were compared with other nonselective NSAIDs. No results from past or ongoing long-term placebo-controlled RCTs that address the cardiovascular risk of traditional NSAIDs have been published. Fortunately, an increasing number of observational studies are being published, and a total of three meta-analyses of epidemiological studies have been published that summarized the available data.2,3 Despite inclusion of different studies, most evidence is common to all three meta-analyses, and the summary risk estimates obtained were extremely similar and compatible with the evidence from clinical trials. Overall, cardiovascular risk seems to vary according to the individual traditional NSAID. Naproxen, a drug with low selectivity for COX2, high affinity for COX1 and an extended plasma half-life, does not seem to increase cardiovascular risk, and

might even be associated with a small reduction in risk of myocardial infarction. By contrast, diclofenac, an NSAID with greater selectivity for COX2 than COX1 and a short plasma halflife, is associated with a 40% increased risk of cardiovascular events—similar to the excess risk observed with selective COX2 inhibitors. This increase is consistent with results from the MEDAL program, which showed no difference in cardiovascular risk between etoricoxib and diclofenac.4 The magnitude of cardiovascular risk, for other nonselective NSAIDs with available data, lies between that of naproxen and diclofenac. Although COX2 selectivity partly seems to explain the increased cardiovascular risk, it is clearly not the only factor involved; other characteristics such as dose, treatment duration, and plasma half-life might also determine the prevalence of thrombotic events. From the clinician’s perspective, new concerns about cardiovascular risk have added some complexity to the task of choosing the optimal NSAID agent for an individual patient. When NSAIDs are prescribed, both the gastrointestinal and cardiovascular safety profiles of these drugs must be taken into account. Unfortunately, information on the dose and treatment-duration effects related to cardiovascular risk is limited for many of the available nonselective NSAIDs. The influence of concomitant low dose aspirin and differences in cardiovascular risk between the doses studied so far are not clear. Hopefully, future well-performed observational studies will help to characterize the cardiovascular safety of traditional NSAIDs.

Acknowledgments The synopsis was written by Jasmine Farsarakis, Associate Editor, Nature Clinical Practice.

Competing interests The authors declared they have no competing interests.

Correspondence CEIFE Almirante, 28 Madrid 28004 Spain [email protected] Received 5 January 2007 Accepted 5 February 2007 Published online XXXX 2007

D P E C T N EC R R F O O C O N R U P

XXX 2007 VOL 3 NO X

References 1 Kearney PM et al. (2006) Do selective cyclooxygenase-2 inhibitors and traditional non-steroidal anti-inflammatory drugs increase the risk of atherothrombosis? Meta-analysis of randomised trials. BMJ 332: 1302–1308 2 Hernández-Díaz S et al. (2006) Non-steroidal antiinflammatory drugs and the risk of acute myocardial infarction. Basic Clin Pharmacol Toxicol 98: 266–274 3 McGettigan P and Henry D (2006) Cardiovascular risk and inhibition of cyclo-oxygenase: a systematic review of the observational studies of selective and nonselective inhibitors of cyclo-oxygenase 2. JAMA 296: 1633–1644 4 Cannon CP et al. (2006) Cardiovascular outcomes with etoricoxib and diclofenac in patients with osteoarthritis and rheumatoid arthritis in the Multinational Etoricoxib and Diclofenac Arthritis Long-term (MEDAL) programme: a randomised comparison. Lancet 368: 1771–1781

LA García Rodríguez is the Director of Epidemiology and A González-Pérez is an Epidemiologist at the Spanish Centre for Pharmacoepidemiologic Research in Madrid, Spain.

www.nature.com/clinicalpractice doi:10.1038/ncprheum0459

PRACTICE POINT Further observational studies are needed to characterize the cardiovascular safety of traditional NSAIDs

NATURE CLINICAL PRACTICE RHEUMATOLOGY 3