Cause-Specific Cardiovascular Risk Associated With Nonsteroidal Antiinflammatory Drugs Among Healthy Individuals

Background— Studies have raised concern on the cardiovascular safety of nonsteroidal antiinflammatory drugs (NSAIDs). We studied safety of NSAID therapy in a nationwide cohort of healthy individuals. Methods and Results— With the use of individual-level linkage of nationwide administrative registers, we identified a cohort of individuals without hospitalizations 5 years before first prescription claim of NSAIDs and without claimed drug prescriptions for selected concomitant medication 2 years previously. The risk of cardiovascular death, a composite of coronary death or nonfatal myocardial infarction, and fatal or nonfatal stroke associated with the use of NSAIDs was estimated by case-crossover and Cox proportional hazard analyses. The entire Danish population age 10 years or more consisted of 4 614 807 individuals on January 1, 1997, of which 2 663 706 (57.8%) claimed at least 1 prescription for NSAIDs during 1997 to 2005. Of these; 1 028 437 individuals were included in the study after applying selection criteria regarding comorbidity and concomitant pharmacotherapy. Use of the nonselective NSAID diclofenac and the selective cyclooxygenase-2 inhibitor rofecoxib was associated with an increased risk of cardiovascular death (odds ratio, 1.91; 95% confidence interval, 1.62 to 2.42; and odds ratio, 1.66; 95% confidence interval, 1.06 to 2.59, respectively), with a dose-dependent increase in risk. There was a trend for increased risk of fatal or nonfatal stroke associated with ibuprofen treatment (odds ratio, 1.29; 95% confidence interval, 1.02 to 1.63), but naproxen was not associated with increased cardiovascular risk (odds ratio for cardiovascular death, 0.84; 95% confidence interval, 0.50 to 1.42). Conclusions— Individual NSAIDs have different degrees of cardiovascular safety, which must be considered when choosing appropriate treatment. In particular, rofecoxib and diclofenac were associated with increased cardiovascular mortality and morbidity and should be used with caution in most individuals, whereas our results suggest that naproxen has a safer cardiovascular risk-profile.