Effect of cyclooxygenase-2 inhibition on renal function in elderly persons receiving a low-salt diet. A randomized, controlled trial.

BACKGROUND: Most nonsteroidal anti-inflammatory drugs (NSAIDs) inhibit both cyclooxygenase-1 (COX-1), whose inhibition is associated with gastrointestinal ulceration, and COX-2, whose inhibition is associated with therapeutic benefits. Although agents that do not produce COX-1 activity may have fewer adverse effects, targeted disruption of the COX-2 allele in mice has resulted in severe renal problems, suggesting that COX-2 inhibition may also produce adverse effects. OBJECTIVE: To determine the effect of rofecoxib, a member of the coxib class of drugs and a specific inhibitor of the COX-2 enzyme, on renal function in elderly patients. DESIGN: A randomized, three-period, single-dose crossover study and a randomized, parallel-group, multiple-dose study. SETTING: Clinical research units. PATIENTS: 75 patients 60 to 80 years of age. INTERVENTION: In the first study, single doses of rofecoxib, 250 mg (about 5-fold to 20-fold the recommended dose); indomethacin, 75 mg; and placebo were administered to 15 patients. In the second study, multiple doses of rofecoxib, 12.5 or 25 mg/d; indomethacin, 50 mg three times daily; or placebo were administered to 60 patients. Patients in both studies received a low-sodium diet MEASUREMENTS: Glomerular filtration rate, creatinine clearance, and urinary and serum sodium and potassium values. RESULTS: Compared with placebo, single doses of rofecoxib and indomethacin decreased the glomerular filtration rate by 0.23 m/s (P < 0.001) and 0.18 mL/s (P = 0.003), respectively. In contrast, respective decreases of 0.14, 0.13, and 0.10 mL/s were observed after multiple doses of rofecoxib, 12.5 mg/d (P = 0.019); rofecoxib, 25 mg (P = 0.029), and indomethacin (P = 0.086) were administered. Changes in creatinine clearance and serum and urinary sodium and potassium were less pronounced. CONCLUSIONS: The effects of COX-2 inhibition on renal function are similar to those observed with nonselective NSAIDs. Thus, COX-2 seems to play an important role in human renal function.

Role of duration of diuretic effect in preventing sodium retention.

OBJECTIVE: To determine whether the duration of diuretic effect at the active nephron site enhances ability to excrete an exogenous salt load. METHODS: We conducted a study that involved eight patients with New York Heart Association class II to III congestive heart failure. In a randomized, crossover manner, each patient received 3.25 mg intravenous bumetanide at 0 hours and again at 6 hours or a loading dose of 0.5 mg bumetanide at 0 hours followed by a continuous infusion of 0.5 mg/hr for 6 hours. Response was followed for 12 hours; a total of 6.5 mg of bumetanide was administered in each arm of the study. Eight hours after dosing began, we administered approximately 80 mEq sodium intravenously and examined its excretion over 4 hours. RESULTS: The percentage of the load excreted was 86% +/- 15% versus 29% +/- 30% for the infusion and bolus regimens, respectively (p = 0.0005). More bumetanide was excreted during the infusion (667 +/- 133 micrograms versus 240 +/- 121 micrograms; p = 0.0002). During the infusion, however, more sodium was excreted relative to amounts of bumetanide, indicating that the efficiency of response was greater during the infusion, 0.10 +/- 0.02 mEq sodium per microgram bumetanide versus 0.07 +/- 0.05 mEq for the bolus (p = 0.0145). CONCLUSIONS: These data support the notions that a long-acting loop diuretic maintains its efficacy and that a longer duration of action facilitates excretion of a sodium load, such as that which might occur during dietary indiscretion.