Long-Term Cardiovascular Risks Associated With NSAID Use After Myocardial Infarction

Richard E. Stewart, MD, FACC , FSCAI

Review

Schjerning Olsen A-M, Fosbol EL, Lindhardsen J, et al. Long-term cardiovascular risk of NSAID use according to time passed after first-time myocardial infarction: a nationwide cohort study. Circulation. 2012;126:1955-1963.

Nonsteroidal anti-inflammatory drugs (NSAIDs) are widely used both in prescription and overthe- counter formulations. The use of these agents for a wide range of indications increases with age. Thus, defining the long-term safety profile of NSAIDs is of major clinical importance.

In the 2007 guideline update, the American Heart Association discouraged NSAID use in patients with established cardiovascular diseases, including coronary artery disease, prior myocardial infarction (MI), and heart failure.1 In the general aging population, however, NSAID use is prevalent and increasing in patients who have a cardiovascular diagnosis. The risks associated with short-term use of NSAIDs in patients who have sustained a firsttime MI are well established.2 Whether the same level of risk is sustained longterm— years after the index event—has been the subject of debate.

Defining the risks of continuous NSAID use years after first-time MI is additionally important, since the overall risk for cardiovascular events and death rapidly declines for all patients during the first year after MI.3 Schjerning Olsen et al4 addressed these concerns in a nationwide cohort study of all adult patients age 30 years and over with

first-time MIs in Denmark from 1997 to 2009, using data obtained from individual- level linkage of hospitalization and pharmacy registries.

Study Details

The study aimed to determine whether the cardiovascular risk associated with using NSAIDs was associated and sustained with time elapsed following first-time MI. The authors utilized the Danish National Patient Registry for all hospital admissions and the Danish Registry of Medicinal Products Statistics for all recorded prescriptions dispensed in Denmark from 1997 to 2009.5,6 A total of 99,187 adult patients who were alive 30 days after discharge for first-time MI (mean age, 69 years; 64% men, 36% women) were studied. Of these patients, 43,608 (44%) were prescribed NSAIDs after the index MI, including the COX- 2 inhibitors celecoxib and rofecoxib, and nonselective agents diclofenac, naproxen, and ibuprofen.

The incident rates of death in the study cohort were determined, as well as composite end points of coronary death or non-fatal recurrent MI associated with NSAID use up to 5 years after study entry. The risk for cardiovascular death was determined by using multivariable adjusted, time-dependent Cox proportional- hazard models. Co-morbidities were defined by patient use of loop diuretics and glucose-lowering agents as proxies for heart disease and diabetes. Age and sex were accounted for in the analysis. There were 36,747 deaths recorded in the study cohort; 28,693 composite end point events of coronary death or non-fatal MI were recorded during the 5-year follow-up. Death rates per 100 person-years were calculated for NSAID use in general, individual NSAIDs used, and patients not treated with NSAIDs. The use of any NSAID following first MI was persistently associated with an increased risk of death after 1 year (hazard ratio [HR], 1.59; confidence interval [CI], 1.49-1.69) and after 5 years (HR, 1.63; CI, 1.52-1.74). The risk of coronary death or non-fatal recurrent MI was also associated with an increased risk at 1 and 5 years (HR, 1.30; CI, 1.22-1.39 and HR, 1.41; CI, 1.28-1.55, respectively). Comorbidities such as the use of loop diuretics and glucose-lowering agents were found to predict use of NSAIDs. The authorsconcluded that the use of NSAIDs is associated with persistently increased coronary event risk, regardless of time elapsed after first-time MI.

References

1. Antman EM, Bennett JS, Taubert KA, et al. Use of nonsteroidal anti-inflammatory drugs: an update for clinicians: a scientific statement from the American Heart Association. Circulation. 2007;115:1634-1642.

2. Schjerning Olsen A-M, Fosbol FL, Gislason GH, et al. Duration of treatment with nonsteroidal anti-inflammatory drugs and impact on risk of death and recurrent infarction in patients with prior infarction: a nationwide cohort study. Circulation. 2011; 123:2226-2235.

3. Rasmussen S, Zwisler AD, Madsen JK, et al. Hospital variation in mortality after first acute myocardial infarction in Denmark from 1995 to 2002: lower short-term and 1-year mortality in higher volume and specialized hospitals. Med Care. 2005;43:970-978.

4. Schjerning Olsen A-M, Fosbol EL, Lindhardsen J, et al. Long-term cardiovascular risk of NSAID use according to time passed after first-time myocardial infarction:

a nationwide cohort study. Circulation. 2012; doi:101.1161/CirculationAHA. 112.112607.

5. Andersen TF, Madsen M, Jorgensen J, Mellemkjoer L, Olson JH. The Danish National Hospital Register: a valuable source of data for modern health sciences. Dan Med Bull. 1999;46:263-268.

6. Gaist, D, Sorensen HT, Hallas J. The Danish prescription registries. Dan Med Bull. 1997;44:445-448.

7. Gislason GH, Jacobsen S, Torp-Pedersen C, et al. Risk of death or reinfarction associated with the use of selective cyclooxygenase- 2 inhibitors and nonselective nonsteroidal anti-inflammatory drugs after acute myocardial infarction. Circulation. 2006;113:2906-2913.

The important issue addressed by the present study by Schjerning Olsen et al is the short- and longterm safety profile of NSAIDs, a widely used class of medications in patients who have sustained a first-time MI. It has not been fully established whether the risk for cardiovascular death and subsequent events would decline in patients taking NSAIDs over time, as seen in the general population after first-time MI. The authors’ conclusion that NSAID use among patients with firsttime MI was associated with persistently increased risk of all-cause mortality and a composite of coronary death and non-fatal recurrent MI lasting at least 5 years has important clinical implications. The study confirms other reports suggesting that NSAIDs do not have a safe treatment window among patients with MI.7

The strength of this study lies in its database, which encompassed the entire country of Denmark through complete registries of hospital admissions and pharmacy records. The complete registration of all residents from Denmark, from which the study cohort was drawn, greatly reduces any selection bias. Also, medication compliance after MI is well documented in this particular study group, so discontinuation of medications such as aspirin in lieu of NSAID use was felt not to be an influencing factor in interpreting the results.

The authors acknowledge that the main limitation of the study was its observational nature. Also, clinical confounders such as heart failure and diabetes were not directly measured and could possibly influence the study results. The nature of the study cohort, however, would tend to mitigate these confounding effects. Furthermore, this contention is supported by the fact that an unmeasured confounder had to elevate the cardiovascular event risk more than 4-fold to explain the risk increase by NSAID use.

The authors noted additional clinical implications of the study, some of which are based on prior studies. It has been noted that individual NSAIDs used for similar indications with different degrees of COX-2 inhibition correlated with risk. This indicates the importance of the NSAID drug effect, and not drug-drug interactions, in explaining the study results. Second, the cardiovascular risk with NSAIDs is dose-dependent, as reported previously.2 Finally, although the study results and conclusions extend to NSAIDS as a class, individual drug analyses confirmed that naproxen was the NSAID with the lowest relative cardiovascular risk, and should be considered in post- MI patients when NSAID use is mandated.7 As a drug class, NSAIDs remain widely used, both as prescribed and over-the-counter products. The study by Schjerning Olsen et al suggests advising caution at all times in using these medications after MI.

References

1. Antman EM, Bennett JS, Taubert KA, et al. Use of nonsteroidal anti-inflammatory drugs: an update for clinicians: a scientific statement from the American Heart Association. Circulation. 2007;115:1634-1642.

2. Schjerning Olsen A-M, Fosbol FL, Gislason GH, et al. Duration of treatment with nonsteroidal anti-inflammatory drugs and impact on risk of death and recurrent infarction in patients with prior infarction: a nationwide cohort study. Circulation. 2011; 123:2226-2235.

3. Rasmussen S, Zwisler AD, Madsen JK, et al. Hospital variation in mortality after first acute myocardial infarction in Denmark from 1995 to 2002: lower short-term and 1-year mortality in higher volume and specialized hospitals. Med Care. 2005;43:970-978.

4. Schjerning Olsen A-M, Fosbol EL, Lindhardsen J, et al. Long-term cardiovascular risk of NSAID use according to time passed after first-time myocardial infarction:

a nationwide cohort study. Circulation. 2012; doi:101.1161/CirculationAHA. 112.112607.

5. Andersen TF, Madsen M, Jorgensen J, Mellemkjoer L, Olson JH. The Danish National Hospital Register: a valuable source of data for modern health sciences. Dan Med Bull. 1999;46:263-268.

6. Gaist, D, Sorensen HT, Hallas J. The Danish prescription registries. Dan Med Bull. 1997;44:445-448.

7. Gislason GH, Jacobsen S, Torp-Pedersen C, et al. Risk of death or reinfarction associated with the use of selective cyclooxygenase- 2 inhibitors and nonselective nonsteroidal anti-inflammatory drugs after acute myocardial infarction. Circulation. 2006;113:2906-2913.

About the Author

Richard E. Stewart, MD, FACC, FSCAI, is a coronary and endovascular interventional cardiologist at The Heart Center of Bakersfield in Bakersfield, CA. He is Clinical Associate Professor of Medicine in the Cardiology Section of St. Louis University Health Science Center in St. Louis. Dr Stewart received his MD from George Washington University in Washington, DC, and completed his Interventional Cardiology Fellowship at the William Beaumont Hospital in Royal Oak, MI, and a Vascular Interventional Fellowship at the Galichia Research Institute in Wichita, KS.