When cyclooxygenase (COX)-2–selective inhibitors (coxibs) first entered the market about 5 years ago, the major concern with regard to cardiovascular side effects related to their potential to increase blood pressure and cause salt and water retention, in a manner similar to conventional (nonselective) nonsteroidal antiinflammatory drugs (NSAIDs). In the short time since, wariness about these side effects has grown into widespread alarm about putative prothrombotic actions and generation of excess major cardiovascular events with these agents. See p 1024 In September 2004, safety findings of the Adenomatous Polyp Prevention on Vioxx (APPROVe) study indicated an increase in risk for myocardial infarction and stroke among subjects randomized to rofecoxib compared with those randomized to placebo. This study had been designed to examine recurrent colonic polyps rather than cardiovascular disease end points and therefore could only be considered hypothesis generating. Subsequent data from a number of observational studies further implicated the drug’s association with arterial thromboembolic disease. As with APPROVe, no single study generated sufficiently robust attestation by itself, but they provided concordant signals, and the accumulating evidence eventually reached a critical mass. As the data on rofecoxib emerged, concerns about other COX-2–selective agents on the market also grew. This was based on the assumption of a “class effect” for an increase in risk of cardiovascular disease related to preferential inhibition of prostacyclin over thromboxane and thus a tendency toward prothrombosis. The results presented by McAdam and colleagues in this issue of Circulation 1 challenge our view with regard to this prostanoid hypothesis and raise new questions about the mechanisms underlying the potential cardiovascular side effects of coxibs and traditional NSAIDs. The authors hypothesized that the enhanced systemic biosynthesis of prostacyclin in smokers was dependent on COX-2 activity. They also sought to determine the functional importance of COX2–derived prostacyclin (PGI2) in limiting platelet activation in smokers in vivo. Contrary to the conventional wisdom of an expected increase in the biosynthesis of TxA2 (related to increased platelet activation via COX-1), the authors demonstrated the opposite finding—ie, that COX-2 contributed to, and a COX-2 inhibitor reduced, TxA2 biosynthesis in smokers.