Home > DARE Reviews > Risk of acute myocardial infarction with...

PubMed Health. A service of the National Library of Medicine, National Institutes of Health.

Database of Abstracts of Reviews of Effects (DARE): Quality-assessed Reviews [Internet]. York (UK): Centre for Reviews and Dissemination (UK); 1995-.

Database of Abstracts of Reviews of Effects (DARE): Quality-assessed Reviews [Internet].

Risk of acute myocardial infarction with nonselective non-steroidal anti-inflammatory drugs: a meta-analysis

G Singh, O Wu, P Langhorne, and R Madhok.

Review published: 2006.

CRD summary

This review evaluated whether non-selective non-steroidal anti-inflammatory drugs increase the risk of acute myocardial infarction. The authors concluded that the evidence suggests a general trend of increased risk, particularly with the use of diclofenac and ibuprofen. Some methodological limitations of the review mean that the reliability of the authors' conclusions is unclear.

Authors' objectives

To determine whether non-selective non-steroidal anti-inflammatory drugs (NSAIDs) increase the risk of acute myocardial infarction (AMI).

Searching

MEDLINE, BIDS and EMBASE were searched from January 1980 to June 2005; the search terms were reported. Experts were contacted, and relevant discussions from the U.S. Food and Drug Administration

advisory panels and the UK National Institute of Clinical Excellence were reviewed. The reference lists of relevant papers and reviews were handsearched.

Study selection

Study designs of evaluations included in the review

Observational studies were eligible for inclusion.

Specific interventions included in the review

Studies of non-selective NSAIDs were eligible for inclusion. The NSAIDs used in the review were diclofenac, ibuprofen, naproxen and other categories, including non-aspirin and non-naproxen. NSAID exposure was reported to be short in most studies, and information on dose was available in under half of the included studies. Reported concomitant medications included aspirin.

Participants included in the review

Studies of current NSAID users compared with non-users or remote users were eligible for inclusion. The included NSAID users were aged between 18 and 101 years. The majority of studies included participants from non-specified populations, with the exception of menopausal women and rheumatoid arthritis patients. The studies were located in Canada, Denmark, the United Kingdom and the USA.

Outcomes assessed in the review

Studies of objectively confirmed AMI were eligible for inclusion. The included outcomes were first hospitalised AMI, hospitalised AMI, first AMI, stroke, death due to coronary heart disease, sudden cardiac death and sudden death.

How were decisions on the relevance of primary studies made?

Two reviewers independently selected studies for the review, and any disagreements were resolved by discussion.

Assessment of study quality

Two independent reviewers assessed study validity, and any disagreements were resolved by discussion. A modified checklist for quantitative studies was used.

Data extraction

Two independent reviewers extracted the data, and any disagreements were resolved by discussion. Relative risks (RRs) or odds ratios (ORs) of AMI associated with exposure to non-selective NSAIDs as a class and to individual NSAIDs (diclofenac, ibuprofen and naproxen) were extracted, along with corresponding 95% confidence intervals (CIs).

Methods of synthesis

How were the studies combined?

The studies were combined in a meta-analysis based on a random-effects model with inverse variance weights. Pooled estimates of RR for AMI and corresponding 95% CIs were computed for exposure to NSAIDs as a class, and for individual NSAIDs. Study-level ORs were pooled with RRs on the assumption that an OR estimated RR when the outcome was rare. A funnel plot was used to assess publication bias.

How were differences between studies investigated?

Heterogeneity of treatment effects was assessed using the chi-squared statistic, and inconsistency was assessed using the I-squared statistic. Sensitivity analyses were conducted to test the robustness of the results and to further explore heterogeneity.

Results of the review

Fourteen observational studies were included in the review: 6 nested case-control studies, 1 matched case-control study, 4 case-control studies and 3 retrospective cohort studies. The number of participants was unclear.

The use of non-selective NSAIDs (as a class) was associated with a statistically significant increase in the occurrence of AMI when compared with non-use or remote use (RR 1.19, 95%CI: 1.08, 1.31, p=0.0006; 13 studies). Heterogeneity was statistically significant (p<0.00001; I-squared 83.8%).

The use of diclofenac was associated with a statistically significant increase in the occurrence of AMI when compared with non-use or remote use (RR 1.38, 95% CI: 1.22, 1.57, p<0.00001; 5 studies). Heterogeneity was statistically non significant, but inconsistency was moderate (I-squared 54%).

The use of ibuprofen was associated with a statistically significant increase in the occurrence of AMI when compared with non-use or remote use (RR 1.11, 95% CI: 1.06, 1.17, p=0.0001; 9 studies). There was no statistical evidence of heterogeneity.

The use of naproxen was not associated with a statistically significant increase in the occurrence of AMI when compared with non-use or remote use (12 studies). Heterogeneity was statistically significant (p=0.01) and inconsistency was moderate (I-squared 54%).

Sensitivity analyses.

The omission of 2 studies of special populations had little impact on the pooled RR for AMI associated with non-selective NSAIDs, or on the heterogeneity of RR estimates. The omission of a study with an extreme value (RR=1.68) reduced the risk of AMI (RR 1.13, 95% CI: 1.07, 1.18, p<0.00001). Heterogeneity remained statistically significant but consistency improved (I-squared 30.7%). The omission of 4 of the 5 studies from the same database did not affect the estimate of overall risk.

Authors' conclusions

The use of non-selective NSAIDs is associated with an increase in AMI risk in comparison with remote or non-use. More specifically, limited evidence suggests that the use of diclofenac or ibuprofen, but not naproxen, is associated with this increased risk.

CRD commentary

This review was supported by clearly defined inclusion criteria and a comprehensive search strategy. Although publication bias was assessed, the results were not reported. The possible omission of unpublished studies might mean that relevant data were missed. Adequate steps were taken to help minimise the potential for bias and error in all phases of the review process. However, the quality of the included studies (and any subsequent synthesis) could not be determined because the results of the

quality assessment were not reported.

It is unclear whether meta-analysis was the most appropriate method of synthesis given the heterogeneity detected in some of the analyses. The authors acknowledged the known limitations

associated with pooling data from observational studies. In addition, they pointed out potential bias associated with studies supported by pharmaceutical company funding. Sensitivity analyses were undertaken to investigate possible sources of heterogeneity, but study features such as type of observational design, duration or quality were not considered. The authors' conclusions reflect the evidence presented but, given the limitations highlighted, their reliability is unclear.

Implications of the review for practice and research

Practice: The authors stated that all NSAIDs should be used at the lowest possible dose for the shortest possible duration. NSAIDs should be used with caution in people with atheromatous vascular disease and should be avoided in people at risk of clinical complications.

Research: The authors stated that further studies of AMI risk associated with the use of non-selective NSAIDs are warranted; ideally, a large randomised controlled trial should be conducted. Larger studies of concomitant aspirin use and studies of NSAID dose and duration are needed.

Bibliographic details

Singh G, Wu O, Langhorne P, Madhok R. Risk of acute myocardial infarction with nonselective non-steroidal anti-inflammatory drugs: a meta-analysis. Arthritis Research and Therapy 2006; 8(5): R153. [PMC free article: PMC1779447] [PubMed: 16995929]

Indexing Status

Subject indexing assigned by NLM

MeSH

Anti-Inflammatory Agents, Non-Steroidal /adverse effects; Humans; Myocardial Infarction /epidemiology /etiology; Risk Factors

AccessionNumber

12007005119

Database entry date

09/08/2008

Record Status

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.

CRD has determined that this article meets the DARE scientific quality criteria for a systematic review.

Copyright © 2014 University of York.

PMID: 16995929