COX-2 inhibitors, nonselective NSAID use increases risk of death, reinfarction in acute MI patients

A meta-analysis of data from national hospital records in Denmark and from the country's national prescription registry showed that the use of selective cyclooxygenase-2 (COX-2) inhibitors in all doses and nonselective non-steroidal anti-inflammatory drugs (NSAIDs) in high doses raised the risk of death in patients who experienced first-time acute myocardial infarction (MI).

Researchers also observed an increased risk for rehospitalization for MI in patients who used either selective COX-2 inhibitors or nonselective NSAIDs.

"Given the additional evidence from randomized trials and other observational studies of selective COX-2 inhibitors and nonselective NSAIDs, these drugs should be used with particular caution in patients with a prior MI," the authors stated.

Among nonselective NSAIDs, ibuprofen (n=10,230) and diclofenac (n=6,172) represented the majority and were separated for analysis from all other NSAIDs (n=7,449). Because the national prescription registry lacked information on prescribed daily dosages, researchers calculated average daily dosages from up to 3 consecutive prescriptions. Rofecoxib and celecoxib were divided into low (≤25 mg and ≤200 mg, respectively) and high (>25 mg and >200 mg) daily dosages. Researcher also separated ibuprofen and diclofenac by dose (≤1,200 mg and <100 mg, respectively, for low dosages). Dosages for other NSAIDs were not divided.

Of the 71,515 patients admitted for first-time MI, 58,432 were alive and were discharged from the hospital, and 21,903 patients claimed at least 1 prescription for a study NSAID. Multivariable Cox proportional hazards models adjusted for age, gender, year of first MI, length of treatment, concomitant medical status, and comorbidity were used to calculate hazard ratios. A subsequent case-crossover analysis employing the cases as their own controls was used to confirm the results of the hazard analysis among patients experiencing an event of interest and to eliminate potential confounders by keeping age, sex, socioeconomic status, and comorbidity fixed.

Between COX-2 inhibitors, rofecoxib had the highest increased risk for death (HR=2.80; 95% CI, 2.41–3.25; P<.0001), while celecoxib (n=2,489) had a hazard ratio of 2.57 (95% CI, 2.15–3.08; P<.0001). Among nonselective NSAIDs, diclofenac had the highest increased risk for death (HR=2.40; 95% CI, 2.09–2.80; P<.0001), followed by ibuprofen (HR=1.50; 95% CI, 1.36–1.67; P<.0001) and other NSAIDs (HR=1.29; 95% CI, 1.16–1.43; P<.0001). For rehospitalization for MI, patients with any use rofecoxib and celecoxib demonstrated an increased risk (HR=1.63; 95% CI, 1.27–2.10; P=.0001 and HR=1.50; 95% CI, 1.10–2.05; P=.01, respectively), as did use of a high daily dose of diclofenac (HR=1.89; 95% CI, 1.40–2.55; P<.0001).

A dose-dependent relationship was observed for rofecoxib (HR=5.26; 95% CI, 3.90–7.09; P<.0001), celecoxib (HR=4.69; 95% CI, 3.58–6.14; P<.0001), ibuprofen (HR=2.20; 95% CI, 1.95–2.48; P<.0001), and diclofenac (HR=4.44; 95% CI, 3.79–5.19; P<.0001), which all raised the risk of death at high dose levels.

Use of the case crossover design confirmed the results from the initial Cox proportional hazards analyses with any use of nonselective NSAIDs and any use of COX-2 inhibitors increasing the risk of death, especially at higher doses. The trend for rehospitalization for MI was more evident among patients with any use of rofecoxib (OR=2.46; 95% CI, 1.42–4.24; P=.001), high daily doses of celecoxib (OR=5.27; 95 CI%, 1.07–25.9; P=.04), any use of ibuprofen (OR=1.32; 95% CI, 1.02–1.72; P=.04), and any use of diclofenac (OR=1.67; 95% CI, 1.15–2.42; P=.007).

The authors said the harmful effects could conceivably be attributed to the class of medications rather than a particular medication because the risks exhibited by the 2 COX-2 inhibitors were similar. Prior observational research had not demonstrated greatly increased cardiovascular risk with celecoxib.

"Because the population in the present study comprised post-MI patients, this might indicate more abrupt effects on a vulnerable cardiovascular system, and the harmful effects of celecoxib may therefore become more apparent than in previous studies," the authors said.

A limitation identified by researchers was that the effects of over-the-counter aspirin use could not be assessed.

The condition that led to the patients' initial use of NSAIDs or COX-2 inhibitors also may have served to raise the risk of events and inflated the apparent risk posed by NSAIDs, according to an accompanying editorial written by Judith Hochman, MD, and Nirav Shah, MD, Cardiovascular Clinical Research Center and Division of General Internal Medicine, New York University School of Medicine. Drs Hochman and Shah suggested that the addition of a placebo or high-dose aspirin group would improve future assessments of the cardiovascular risks associated with the agents.

"This latest estimate of the cardiovascular risk of NSAIDs provides cause for concern," Drs Hochman and Shah stated. "If even relatively short-term (ie, 1-month) use is associated with relatively large risk for death and reinfarction in those with prior MI, then widespread use of such agents, especially in light of fewer treatment options with COX-2 withdrawals, may be resulting in substantial morbidity."

Drs Hochman and Shah indicated that a future noninferiority study will compare naproxen, ibuprofen, and celecoxib among 21,000 patients.

SOURCES Gislason GH, Jacobsen S, Rasmussen JN, et al. Risk of death or reinfarction associated with the use of selective cyclooxygenase-2 inhibitors and nonselective nonsteroidal antiinflammatory drugs after acute myocardial infarction. Circulation. 2006;113:2906–2913.

Hochman JS, Shah, NR. What price pain relief? Circulation. 2006;113:2868–2870.