Inhibition of clinical benefits of aspirin on first myocardial infarction by nonsteroidal anti-inflammatory drugs Kruth T, Glynn RJ, Walker AN, et al. Circulation 2003;108:1191-5

AbstractConclusionThere may be an increased rate of myocardial infarction (MI) in apparently healthy male physicians taking prophylactic aspirin therapy who also use nonsteroidal anti-inflammatory agents (NSAIDs) on a regular basis, compared with those taking aspirin therapy with less use of NSAIDs. Regular use of NSAIDs not specific to the cyclooxygenase-2 (COX-2) coenzyme may impede the beneficial cardiovascular effects of aspirin.SummaryThis is a subgroup analysis of the Physician's Health Study (PHS). In the PHS, 23,071 apparently healthy male physicians between 40 and 84 years of age at entry and without a history of cardiovascular disease or other major illness were randomized to receive either 325 mg of aspirin every other day, β-carotene every day, both active agents, or both placebos. The study was stopped in January 1998, because of the emergence of a significant beneficial effect of aspirin on the development of first myocardial infarction. Aspirin provided a 44% risk reduction in myocardial infarction in the participants in the study (P < .00001).Aspirin is thought to prevent ischemic events by irreversible inhibition of cyclooxygenase-1 (COX-1). NSAIDs exert their effects primarily by inhibition of COX-2 and subsequent reduced prostaglandin production. Many NSAIDs, however, also inhibit COX-1. Aspirin and NSAIDs share a common binding site on COX-1. NSAIDs may compete for binding sites with aspirin on the COX-1 isoenzyme, raising the possibility that NSAIDs may obviate the beneficial cardiovascular effects of aspirin.In the PHS, data were recorded on NSAID use. In this subgroup analysis the authors found that participants in the PHS who used NSAIDs from 1 to 59 days per year did not have an increased risk for MI (relative risk [RR], 1.21; 95% confidence interval [CI], 0.78-1.87). However, participants in the study who were randomized to aspirin and who used NSAIDs more than 60 days per year had an increased relative risk for first MI, compared with those who did not use NSAIDs (RR, 2.86; 95% CI, 1.25-6.56).CommentPost hoc analysis such as that presented in this study provides the basis and rationale for design of proper prospective trials. Post hoc analyses should not be regarded as definitive studies. The authors are correct in urging caution in interpreting the implications of their data