Cardiorenal risk of celecoxib compared with naproxen or ibuprofen in arthritis patients: insights from the PRECISION trial

Aims Non-steroidal anti-inflammatory drugs (NSAIDs) are among the most frequently used drugs, both prescribed and over the counter. The long-term cardiovascular safety of NSAIDs in patients with arthritis has engendered controversy. Concerns remain regarding the relative incidence and severity of adverse cardiorenal effects, particularly in arthritis patients with established cardiovascular (CV) disease or risk factors for disease as illustrated by the PRECISION (Prospective Randomized Evaluation of Celecoxib Integrated Safety vs. Ibuprofen Or Naproxen) trial participants (NCT00346216). We further investigated whether the selective COX-2 Inhibitor celecoxib has a superior cardiorenal safety profile compared with ibuprofen or naproxen in the PRECISION population. Methods and results Twenty-four thousand eighty-one patients who required NSAIDs for osteoarthritis or rheumatoid arthritis (RA) and had increased CV risk randomly received celecoxib, ibuprofen, or naproxen. The current pre-specified secondary analysis assessed the incidence, severity, and NSAID-related risk of the pre-specified composite cardiorenal outcome (adjudicated renal event, hospitalization for congestive heart failure, or hospitalization for hypertension) in the intention-to-treat (ITT) population. An on-treatment analysis assessed safety in those taking the study medication. Following a mean treatment duration of 20.3 +/- 16.0 months and a mean follow-up of 34.1 +/- 13.4 months, the primary cardiorenal composite outcome occurred in 423 patients (1.76%) in the ITT population. Of these 423 patients, 118 (28%) were in the celecoxib, 166 (39%) in the ibuprofen, and 139 (33%) in the naproxen group. In a multivariable Cox regression model adjusted for independent clinical variables, celecoxib showed a significantly lower risk compared with ibuprofen [hazard ratio (HR) 0.67, confidence interval (CI) 0.53-0.85, P = 0.001) and a trend to lower risk compared with naproxen (HR 0.79, CI 0.61-1.00, P = 0.058). In the ITT analysis, clinically significant renal events occurred in 220 patients with events rates of 0.71%, 1.14%, and 0.89% for celecoxib, ibuprofen, and naproxen, respectively (P = 0.052), while in the on-treatment analysis the rates were 0.52%, 0.91%, and 0.78% (P < 0.001). Conclusion In the current era, long-term NSAID use was associated with few cardiorenal events in arthritis patients. At the doses studied, celecoxib displayed fewer renal events and hence more favourable cardiovascular safety compared with ibuprofen or naproxen. These results have considerable clinical implications for practitioners managing individuals with chronic arthritis pain and high risk of impaired renal function and/or heart failure. Clinical Trial Registration: NCT00346216