Dispensed selective and nonselective nonsteroidal anti-inflammatory drugs and the risk of moderate to severe hyperkalemia: a nested case-control study.

BACKGROUND: It is not known whether nonselective and cyclooxygenase 2 (COX-2) selective nonsteroidal anti-inflammatory drugs (NSAIDs) differ in terms of hyperkalemia risk. The aim of this study was to compare the differential risk of hyperkalemia associated with various NSAIDs. STUDY
DESIGN: Nested case-control study. SETTING & PARTICIPANTS: New NSAID users receiving care in the US Department of Veterans Affairs (VA) health care system from October 2000 to September 2006. PREDICTOR OF FACTOR: Different NSAIDs ordered by COX-2 selectivity. OUTCOMES & MEASUREMENTS: Risk of hyperkalemia (defined as first serum potassium value ≥6.0 mEq/L) was estimated with a multivariate conditional logistic model adjusting for age, race, morbidities, medications, and contrast media.
RESULTS: We identified 18,326 cases of hyperkalemia and 355,106 matched controls. Risk of hyperkalemia did not differ in patients using a single NSAID (adjusted OR, 1.03; 95% CI, 0.98-1.08) or 2 or more NSAIDs (OR, 1.16; 95% CI, 0.96-1.41) compared with patients no longer using an NSAID. However, risk varied by specific NSAID and elevated risks were found for use of rofecoxib, celecoxib, diclofenac, and indomethacin, but not meloxicam, etodolac, piroxicam, sulindac, ibuprofen, naproxen, and ketorolac. Interactions were found between NSAID use and exposure to renin-angiotensin blockers and contrast media. LIMITATIONS: Some NSAIDs are available over the counter and women are under-represented in the VA population.
CONCLUSIONS: NSAIDs may not by themselves carry a higher risk of moderate to severe hyperkalemia. Certain NSAIDs may increase the risk of hyperkalemia, but it is not related to COX-2 selectivity of the NSAID and may depend on concurrent exposure to other agents.