Meta-Analysis of Cox-2 Inhibitors and Cardiovascular Outcomes

META-ANALYSIS OF COX-2 INHIBITORS AND CARDIOVASCULAR OUTCOMES

Meta-analysis of clinical trials involving COX-2 inhibitors and cardiovascular outcomes conducted by investigators not associated with the pharmaceutical manufacturers were published after the market with-drawal of rofecoxib. Jüni and colleagues abstracted reported frequency of fatal or non-fatal myocardial infarction from 18 clinical trials of rofecoxib among patients with rheumatoid arthritis, osteoarthritis, or back pain (Jüni et al., 2004). Rofecoxib dosage ranged from 12.5 to 50 mg per day and treatment duration ranged from 4 to 56 weeks. The comparison groups received either placebo or a non-selective NSAID. Approximately a third of the 25 573 patients in the 18 trials were enrolled in VIGOR. In 10 of the 18 trials, myocardial infarction was more common among the rofecoxib arm, but the difference only reached statis-tical significance in VIGOR. The combined odds ratio for the rofecoxib versus non-selective NSAID or placebo comparison was 2.24 (95% CI, 1.24–4.02).

Kearney reviewed 138 randomized trials of either a COX-2 inhibitor versus placebo or a COX-2 inhibitor versus non-selective NSAIDs and evalu-ated a range of cardiovascular end points, including vascular events, myocardial infarction, stroke, and vascular death (Kearney et al., 2006). For all COX-2 inhibitors (celecoxib, etoricoxib, lumiracoxib, rofe-coxib, and valdecoxib) as a group and compared with placebo, relative risk of developing serious vascu-lar events was 1.42 (95% CI, 1.13–1.78). Heteroge-neous results were observed when COX-2 inhibitors were compared with different non-selective NSAIDs. COX-2 inhibitors were associated with increased vascular risk when compared with naproxen (relative risk 1.57; 95% CI, 1.21–2.03) and were not associ-ated with increased vascular risk when compared with non-naproxen NSAIDs (relative risk 0.88; 95% CI, 0.69–1.12).

For other adverse cardiovascular outcomes, Aw and colleagues reviewed 19 clinical trials involving COX-2 inhibitors and reported that use of COX-2 inhibitors was associated with increased blood pressure when compared with placebo or non-selective NSAIDs (Aw et al., 2005). The effect on blood pressure was more pronounced among rofecoxib users than among cele-coxib users. Zhang and colleagues reviewed 114 clin-ical trials involving COX-2 inhibitors and found that rofecoxib use was associated with increased risk of arrhythmia, peripheral edema, hypertension, and renal dysfunction but celecoxib use was not associated with increased risk of these events (Zhang, Ding and Song, 2006).