Selective cyclo-oxygenase-2 inhibitors and myocardial infarction: how strong is the link?

There are concerns that selective cyclo-oxygenase (COX)-2 inhibitors may be prothrombotic and increase the risk of myocardial infarction. This has largely arisen because of an unexpected finding of a higher rate of myocardial infarction in patients receiving rofecoxib compared with patients receiving naproxen in a study of gastrointestinal toxicity. The results of this study, a similar study of celecoxib versus ibuprofen or diclofenac, and data obtained from a meta-analysis of aspirin (acetylsalicylic acid) primary prevention trials suggest that differences in the rates of myocardial infarction between rofecoxib and naproxen may have been due to an unexpectedly low rate of myocardial infarction in patients receiving naproxen. However, population surveillance data also suggest that rofecoxib may be associated with a greater risk of myocardial infarction than celecoxib and certain nonselective nonsteroidal anti-inflammatory drugs. The magnitude of this increase in risk, if real, is uncertain but it is likely to be relatively small in patients for whom cardiovascular prophylaxis with aspirin is not indicated. Patients who require nonsteroidal anti-inflammatory therapy for arthritis and who are at high risk of cardiovascular disease should receive aspirin, probably in conjunction with selective COX-2 inhibitor therapy, as the risk of gastrointestinal ulceration may be lower than for aspirin plus a nonselective nonsteroidal anti-inflammatory drug. In patients who do not require aspirin for the prevention of cardiovascular events, the lower risk of gastrointestinal ulceration associated with COX-2 inhibitor compared with non-selective nonsteroidal anti-inflammatory drugs would be expected to outweigh any increase in the risk of myocardial infarction, if one exists.