Prostaglandins (PGs) are biologically active lipids derived from the cyclooxygenase (COX)-mediated metabolism of arachidonic acid. They are constitutively produced in certain tissues (eg, brain, gut, and kidney), and their synthesis is increased at sites of inflammation. Prostanoids function as important mediators of inflammation and modulate a variety of physiological processes, including maintenance of gastric mucosal integrity, renal hemodynamic regulation, renin synthesis and release, and tubular reabsorption of salt and water.1 Cyclooxygenase, or PG synthase G2/H2, is the rate-limiting enzyme responsible for the initial conversion of arachidonic acid to PGG2 and subsequently to PGH2. PGH2 is then metabolized by tissue-specific isomerases to produce prostaglandins and thromboxanes. There are 2 distinct isoforms of cyclooxygenase, COX-1 and COX-2, which share 66% homology in amino acid sequence 2–4 but have different patterns of expression and regulation. COX-1, traditionally termed the “constitutive” enzyme, is widely distributed in tissues, and its level of activity is not dynamically regulated. COX-2, the glucocorticoid-sensitive “inducible” enzyme, is more restricted in its basal expression and is upregulated in response to inflammation, resulting in increased prostanoid production at the site of inflammation. NSAIDs, which block both isoforms COX-1 and COX-2, have been widely used in the treatment of inflammatory conditions. However, the adverse effects of NSAIDs, especially gastrointestinal toxicity, have limited their long-term use in clinical settings. The hypothesis that NSAID-induced gastrointestinal toxicity was related to the inhibition of gastric COX-1, whereas the anti-inflammatory properties caused by COX-2 inhibition5 led to the development of the new antiinflammatory agents, the coxibs, which were designed to inhibit COX-2 selectively. It was anticipated that these selective COX-2 inhibitors would be as effective as the nonselective NSAIDs in the treatment of inflammatory diseases but would be better-tolerated with fewer gastrointestinal side effects. Rofecoxib (Vioxx) and celecoxib (Celebrex) were the first COX-2–selective inhibitors to be marketed as effective antiinflammatory agents without serious gastrointestinal toxicity, based on the results of the VIoxx Gastrointestinal Outcomes Research (VIGOR) trial6 and the Celecoxib Long term Arthritis Safety Study (CLASS) trial, 7 respectively. The VIGOR trial reported that patients with rheumatoid arthritis receiving rofecoxib (50 mg daily) had fewer gastrointestinal events compared with those taking naproxen (500 mg twice daily) (2.1% versus 4.5%; P0.001), but there was similar efficacy between the drugs. However, the incidence of myocardial infarction was increased by a factor of 5 in the rofecoxib-treated group compared with the naproxen group. In the CLASS trial, celecoxib (400 mg twice daily) was compared with ibuprofen (800 mg thrice daily) or diclofenac (75 mg twice daily) in patients with osteoarthritis or rheumatoid arthritis. Celecoxib had a lower gastrointestinal side effect profile compared with other NSAIDs, and there was no difference in the incidence of cardiovascular events between celecoxib and NSAIDs regardless of aspirin use. Questions regarding the cardiovascular risk of the various coxibs remained, and on September 30, 2004 rofecoxib was withdrawn from the market based on the data from the Adenomatous Polyp PRevention On Vioxx (APPROVe) study. 8 This trial was prematurely discontinued after it was discovered that 3.5% of the patients treated with rofecoxib (25 mg once daily) had serious thromboembolic adverse events compared with 1.9% of patients in the placebo group (P0.001).8,9 The