Background Safety data from clinical trials have raised concerns over the cardiovascular (CV) safety of non-steroidal anti-inflammatory drug (NSAID) use in general. Despite the widespread use of NSAIDs in RA patients, the few reports available from RA populations have not indicated an increased CV risk associated with commonly used NSAIDs. This suggests that the benefits of NSAID use (eg. improved mobility and pain relief) to some extent may balance the adverse CV effects observed in non-RA individuals.
Objectives To evaluate the CV risk associated with use of several NSAIDs in RA patients and controls.
Methods The study was designed as a register-based, longitudinal, nationwide cohort study in the period 1997 to 2009. Data were gathered from in- and out-hospital records (diagnoses), pharmacy records (prescriptions) and several civil registers (vital status, emigration and annual income). RA patients were identified by the combination of RA diagnoses and DMARD treatment and matched with 4 randomly selected controls based on age and sex. Individuals with prior myocardial infarction (MI) and stroke were excluded. Overall NSAID exposure and exposure to distinct NSAIDs were determined from redeemed prescriptions. Outcome of interest was the combined endpoint of myocardial infarction, stroke and/or CV mortality (primary cause of death). Risk was assessed by fitting Cox regression models stratified by RA status.
Results 17 320 RA patients were included (71% females, mean age 59 years) and 69 280 matched control. 6 284 events occurred during follow-up (801 during NSAID use). RA patients were exposed to NSAID in 25% of follow-up compared to 5% in the control group. The overall risk associated with NSAID use was significantly lower in RA patients than in controls (hazard ratio 1.21 vs. 1.49, p<0.001). This pattern was also generally present when examining the individual NSAIDs (see table for details), and the risk associated with overall NSAID use in RA patients seemed primarily driven by rofecoxib and diclofenac treatment with little or no risk associated with the other NSAIDs examined. At low and moderate daily dosages, most NSAIDs were associated with no or low risk of adverse CV event in RA patients.
Conclusions This study found that NSAID use in RA patients is frequent and may be associated with an increased risk of CV disease. However, the magnitude of such risks is significantly lower than non-RA individuals and differed among NSAIDs. The results support a strongly individualised approach to NSAID treatment in RA patients based on the need for pain relief, the individual characteristics of the RA patient and the safety profile of the various NSAIDs available.
Disclosure of Interest None Declared