Background Nonsteroidal anti-inflammatory drugs (NSAIDs) can affect sodium reabsorption by decreasing the synthesis of renal prostaglandins. Since COX-2 is constitutively expressed in the kidney, it is hypothesised that the effects of COX-2 inhibitors on sodium handling would be similar to non-selective NSAIDs. It is hypothesised moreover that the effects of the two COX-2 inhibitors on sodium handling would be similar.

Objectives This study evaluated the effects of two COX-2 inhibitors [rofecoxib, 25 mg QD and celecoxib, 200 mg BID], a nonselective NSAID [naproxen, 500 mg BID], and placebo on blood pressure (BP) and urinary sodium excretion during a 2-week in-house administration.

Methods 67 healthy elderly (60 to 80 years) subjects participated in a double-blind, placebo-controlled, parallel-group study. Subjects received a weight-maintaining isocaloric diet (200 mEq sodium, 0.8 g/kg protein, 80 to 120 mEq potassium daily), beginning 8–13 days prior to the first dose of study drug. After attaining sodium balance [based on stable weight (within 0.5 kg) and 24-hour urinary sodium between 180 to 220 mEq on 2 consecutive days], subjects were randomised to treatment. Daily 24-hour urine collections were obtained. Daily BP measurements were taken at 8AM and 8PM, and more often (8 AM, Noon, 4 PM, and 8 PM) on Days -1, 1, 7, and 14.

Results Least-Squares mean changes from baseline (SE) for average daily BP on Day 14 (prespecified primary endpoint for BP) are shown in the Table 1 below. There was no significant difference between the active treatment groups in daily average sodium excretion during the first 3 days or over 2 weeks of treatment (p > 0.05). No incidence of peripheral oedema occurred during the 2 week treatment. Two subjects experienced elevated systolic BP during the course of the trial (one on each COX-2 inhibitor).

Conclusion These results, obtained under well-controlled conditions for two weeks, suggest that the COX-2 selective agents, rofecoxib and celecoxib, have similar effects on BP and urinary sodium excretion; their effect is similar also to that of non-selective NSAIDs.