The effects of short-term piroxicam (P) treatment on natriuresis and kaliuresis were evaluated in 22 diuretic-treated hypertensives and 8 healthy subjects. Each hypertensive patient participated in three experimental periods, i.e., P 20 mg q.d. for 3 days and two control phases. Each healthy volunteer completed the six treatment phases, i.e., P 20 mg q.d. for 3 days, furosemide 40 mg orally in a single dose, P 20 q.d. for 3 days with furosemide 40 mg on the third day and three control phases. In healthy volunteers, P induced no change in 24-hour natriuresis and kaliuresis, but it reduced furosemide-stimulated natriuresis. In hypertensives with creatinine clearance (CCr) greater than 60 ml/min, P had no anti-natriuretic and anti-kaliuretic effects. In contrast, P induced a significant reduction in 24-hour natriuresis and kaliuresis in those with CCr less than 60 ml/min. Physicians should be aware of the potential risk of P-induced sodium retention and hyperkalemia, especially in patients with renal insufficiency, even though P was administered for only 3 days.