Anti-inflammatory therapy and cardiovascular risk: new insights

Abstract

Nonsteroidal anti-inflammatory drugs (NSAIDs) have been the predominant treatment for inflammatory conditions and chronic pain for nearly 40 years. However, when taken on a regular basis, anti-inflammatory agents are associated with a high risk of adverse events involving the upper gastrointestinal tract. It is estimated that approximately 1.1% of those taking NSAIDs develop significant gastrointestinal lesions (bleeding and gastric perforation), for which hospital recovery is necessary [1]. The inhibition of cyclooxygenase (COX) by NSAIDs produces both their therapeutic and their adverse events. The distinction between the enzymatic properties of COX-1 and COX-2 has led to the inception of selective COX-2 antiinflammatory agents (coxibs). The development and commercialization of these agents have led researchers to believe that problems with safety have been resolved, and that prolonged use of these new anti-inflammatory agents would allow easier control of many chronic pathologies, especially rheumatological disorders and arthritis. However, during their clinical development, an unexpectedly high incidence of cardiovascular events was observed. Clinical trial data first demonstrating increased cardiovascular risk with the use of these drugs came from the VIoxx Gastrointestinal Outcomes Research (VIGOR) trial that compared the gastrointestinal tolerability of rofecoxib (50 mg/day) with naproxen (1000 mg/day) in patients with rheumatoid arthritis [2]. The study demonstrated that rofecoxib was associated with a four-times greater risk of myocardial infarction than naproxen. This was attributed initially to a protective effect of naproxen [2], rather than to a specific effect of rofecoxib [3–6]. The problem became more evident in the Adenomatous Polyp PRevention On Vioxx (APPROVe) study carried out in patients with a history of colon adenomas [7]; this trial was interrupted when it was found that rofecoxib was associated with a two-fold higher risk of cardiovascular events than placebo (relative risk [RR] 1.92, confidence interval [CI] 1.19–3.11). In September 2004, Merck Sharpe & Dohme announced the voluntary worldwide withdrawal of Vioxx. Following publication of the APPROVe trial, other data from randomized, controlled clinical trials, as well as epidemiological data, became available. These studies suggested that increased cardiovascular risk is not a problem specific to rofecoxib, but is common to all analogous molecules. In the Adenoma Prevention with Celecoxib (APC) study, conducted in patients with adenomatous colorectal polyps, celecoxib (200 or 400 mg twice daily) was associated with a cardiovascular risk almost three-times higher than that of placebo (RR 2.8, CI 1.3–6.3) [8]. An increased cardiovascular risk, with respect to placebo, was also seen with valdecoxib (and its precursor parecoxib) in patients undergoing a coronary bypass [9,10]. It is likely that, in similar studies, without evidence of increased cardiovascular risk, this was actually due to insufficient statistical power (inadequate number, inclusion of a lowrisk population and low number of adverse events overall) more than intrinsic properties of that particular coxib!