Conventional nonsteroidal anti-inflammatory drugs inhibit both cyclooxygenase (Cox) isoforms (Cox-1 and Cox-2) and may be associated with nephrotoxicity. The present study was undertaken to assess the renal effects of the specific Cox-2 inhibitor, MK-966. Healthy older adults (n = 36) were admitted to a clinical research unit, placed on a fixed sodium intake, and randomized under double-blind conditions to receive the specific Cox-2 inhibitor, MK-966 (50 mg every day), a nonspecific Cox-1/Cox-2 inhibitor, indomethacin (50 mg t.i.d.), or placebo for 2 weeks. All treatments were well tolerated. Both active regimens were associated with a transient but significant decline in urinary sodium excretion during the first 72 h of treatment. Blood pressure and body weight did not change significantly in any group. The glomerular filtration rate (GFR) was decreased by indomethacin but was not changed significantly by MK-966 treatment. Thromboxane biosynthesis by platelets was inhibited by indomethacin only. The urinary excretion of the prostacyclin metabolite 2,3-dinor-6-keto prostaglandin F1alpha was decreased by both MK-966 and indomethacin and was unchanged by placebo. Cox-2 may play a role in the systemic biosynthesis of prostacyclin in healthy humans. Selective inhibition of Cox-2 by MK-966 caused a clinically insignificant and transient retention of sodium, but no depression of GFR. Inhibition of both Cox isoforms by indomethacin caused transient sodium retention and a decline in GFR. Our data suggest that acute sodium retention by nonsteroidal anti-inflammatory drugs in healthy elderly subjects is mediated by the inhibition of Cox-2, whereas depression of GFR is due to inhibition of Cox-1.