OP0024 Concomitant Aspirin Use Reduces the Risk of Acute Myocardial Infarction (AMI) in Users of Cyclooxygenase-2 Selective and Some Non-selective NSAIDs
Sing, Graham, Wang, Mithal, Triadafilopoulos.
An increased risk of acute myocardial infarction (AMI) has been associated with the use of COX-2 selective NSAIDS. In this study, Singh et al examine the use of concomitant aspirin and selective and nonselective cyclooxygenase (COX) inhibitors on AMI using the California Medicaid database.
Methods: Patients with RA or OA who were treated with a COX-2 or non-selective NSAID between January 1999 and June 2004 were assembled. Cases of AMI were identified, and age-, gender- and time-matched to controls. Current exposure to COX-2 or nonselective NSAIDs was compared to remote exposure.
Results: 15,343 cases of AMI (8% fatal) were noted accounting for 2,356,885 person years of follow up. The table below represents the adjusted rate ratios (95% CI) of acute MI as a function of current use of the various NSAIDS with and without concomitant aspirin.
|Without Aspirin||With Aspirin|
|Rofecoxib||1.31 (1.2-1.43)||1.03 (0.86-1.24)|
|Celecoxib||1.12 (1.04-1.19)||0.88 (0.76-1.02)|
|Ibuprofen||1.08 (0.97-1.19)||1.20 (0.94-1.51)|
|Indomethacin||1.65 (1.27-2.15)||1.21 (0.65-2.27)|
|Meloxicam||1.52 (1.14-2.03)||0.53 (0.26-1.10)|
|Sulindac||1.47 (1.03-2.11)||0.77 (0.32-1.85)|
Conclusion: An increased risk of AMI was observed with rofecoxib, celecoxib, indomethacin, meloxicam and sulindac but not ibuprofen. Concomitant aspirin completely abrogated the increased risk of AMI associated with current use of rofecoxib, celecoxib, meloxicam and sulindac. Aspirin incompletely reduced the risk associated with indomethacin, and had no effect on ibuprofen associated risk. These results suggest that aspirin inhibition of thromboxane may balance the selective COX-2 inhibition of prostacyclin formation. No data on GI toxicity between the groups were presented.
Editorial Comments: This study supports the current notion that not only COX-selective, but also some nonselective, NSAIDs raise the risk of acute MI. Interestingly, ibuprofen did not raise the risk. It is curious that the investigators do not report the effect of naproxen in this data set. Aspirin reduced the risk of AMI significantly in the COX-2 users but was less predictable in the non-selective NSAID users. This may relate to the fact that nonselective NSAIDs compete for the same COX binding pocket as aspirin, thus preventing aspirin from binding. Differences in risk ratios for MIs may also relate to the variable effect that an individual NSAID has on blood pressure, some causing elevation of blood pressure while others have little or no effect. This study did not evaluate the risk of NSAID associated GI events, and the potential for concomitant aspirin to undo the GI protective effect of COX-2 selective NSAIDs. As always, one must weigh the potential GI protective effect against the potential cardiovascular risk in each individual patient.