Use of nonsteroidal anti-inflammatory drugs in patients with cardiovascular disease: a cautionary tale

Abstract

Nonsteroidal anti-inflammatory drugs (NSAIDs) inhibit production of prostaglandins by acting on cyclooxygenase (COX) isoenzymes 1 and 2. Nonselective NSAIDs inhibit both COX 1 and 2 isoenzymes (eg, ibuprofen and naproxen). Selective NSAIDs act on COX-1 (eg, aspirin) or COX-2 (eg, celecoxib) isoenzymes, respectively. Prostaglandins are produced in platelets and gastric mucosal cells through constitutively expressed COX-1 isoenzyme. They are involved in the regulation of hemostasis, functional integrity of the gastrointestinal and renal tracts, platelet function, and macrophage differentiation. Inhibition of COX-1 isoenzymes impedes platelet aggregation, impairs maintenance of protective gastric mucosal barrier, and affects renal function. Prostaglandin production in inflamed tissue results from de novo induction of COX-2 expression by inflammatory cytokines and other noxious stimuli. Thus, COX-2 isoenzyme inhibition either selectively or nonselectively helps in reducing inflammation in the setting of musculoskeletal disorders. Safety and efficacy of NSAIDs are related to their relative actions on COX-1 or COX-2 inhibition. Given the multisystem (gastrointestinal, hematopoietic, and renal) adverse effect profile of COX-1 inhibition, formulation of NSAIDs with relative COX-2 selectivity became a highly desirable target during the 90's. However, studies in the first half of this decade revealed adverse effects of COX-2 inhibition on the cardiovascular system, including increased risks of myocardial infarction, exacerbation of stable congestive heart failure, and worsening high blood pressure. Randomized trials and meta-analyses confirmed these findings, which led to withdrawal of some of the COX-2 inhibitors from the market by the federal Food and Drug Administration a few years ago. Here, we review the effects of COX-2 isoenzyme inhibitors on the cardiovascular system to provide a safe strategy for prescribing these agents in patients with existing cardiovascular disease. We did not find adequate long-term randomized controlled trials appropriately powered to evaluate cardiovascular outcomes. Potentially, all NSAIDs possess a fair risk of adverse effects on gastrointestinal, cardiovascular, and renal systems. Until more evidence for safety via randomized trials is available, we recommend caution in prescribing COX-1 and 2 inhibitors for musculoskeletal disorders in patients with existing gastrointestinal or cardiovascular conditions.