The Cox-2 Controversy Continues

In 2001, a report published in the Journal of the American Medical Association suggested that the new, and more gastrointestinally friendly, Cox-2 inhibitors, may place patients at higher risk of myocardial infarction.1 A retrospective analysis of the VIGOR study found that, in patients taking rofecoxib, the incidence of myocardial infarction increased five-fold when compared with patients who received a traditional NSAID, naproxen.

However, the analysis could not determine the reason for this increase. Two mechanisms were proposed to explain the results. The first is that non-selective NSAIDs may be cardioprotective because of their inhibitory effects on thromboxane A2 (TxA2), which causes platelet aggregation. The second is that Cox-2 inhibitors could have deleterious cardiovascular effects because they do not block TxA2 but selectively inhibit the beneficial cardiovascular effects of prostacyclin (PGI2), which is a potent inhibitor of platelet activation by all recognized agonists and a vasodilator.

A recent study on mice has contributed to the discussion, without providing any definitive answers.2 To determine which of the two hypotheses might account for the observed cardiovascular effects, the researchers generated mice with disordered expression of receptors--both PGI2 and TxA2 activate G-protein coupled receptors, the prostacyclin receptor (IP) and the thromboxane receptor (TP), respectively--and monitored the response to vascular injury.

Mice lacking the PGI2 receptor, a model for the clinical effect of taking Cox-2 inhibitors, had an enhanced vascular response to injury and showed increased TxA2 formation and platelet activation. This enhanced response was cancelled out in mice lacking both TxA2 and PGI2 receptors. These data establish that endogenous PGI2 modulates the cardiovascular actions of TxA2 in vitro, an interplay that may be relevant to the cardiovascular effects of selective Cox-2 inhibitors, which depress PGI2 without coincidental inhibition of TxA2 formed by Cox-1 in platelets.

The results of the mouse study are very preliminary, and what occurs in the mouse, is not necessarily relevant to the man. There are also a number of other criticisms of the JAMA study. It has been argued that the dosages used in the VIGOR trial were almost double the normal daily dosage, so even if rofecoxib has a prothrombotic effect, it may be dose-dependent. The increase in mortality in the CLASS trial, while higher in the celecoxib group, resulted from accidents, infection and cancer, and was not related to the use of celecoxib. Unless presented with firm research to the contrary, some are suggesting that Cox-2 selective inhibitors should be considered safe for arthritis patients at risk of gastrointestinal events; however, physicians should take a cautious approach when prescribing rofecoxib to patients with a history of cardiovascular disease.


  1. Mukherjee D, Nissen SE and Topol EJ. Risk of cardiovascular events associated with selective Cox-2 inhibitors. JAMA 2001; 286:954-9.
  2. Cheng Y, Austin SC, Rocca B, et al. Role of prostacylcin in the cardiovascular response to thromboxane A2. Science 2002; 296:539-41.