Peritoneal dialysis (PD) is still widely used for acute renal failure (ARF) in developing countries despite concerns about its inadequacy. Continuous PD has been evaluated in ARF by analyzing the resolution of metabolic abnormality and normalization of plasma pH, bicarbonate, and potassium. A prospective study was performed on 30 ARF patients who were assigned to high-dose continuous PD (Kt/V = 0.65 per session) via a flexible catheter (Tenckhoff) and automated PD with a cycler. Fluid removal, pH and metabolic control, protein loss, and patient outcome were evaluated. Patients received 236 continuous PD sessions; 76% were admitted to ICUs. APACHE II score was 32.2+/-8.65. BUN concentrations stabilized after 3 sessions, creatinine after 4, and bicarbonate and pH after 2. Fluid removal was 2.1+/-0.62 L/day. Creatinine and urea clearances were 15.8+/-4.16 and 17.3+/-5.01 mL/minute respectively. Normalized creatinine clearance and urea Kt/V values were 110.6+/-22.5 L/week/1.73 m(2) body surface area and 3.8+/-0.6 respectively. Solute reduction index was 41%+/-6.5% per session. Serum albumin values remained stable in spite of considerable protein losses (median 21.7 g/day, interquartile range 9.1-29.8 g/day). Regarding ARF outcome, 23% of patients presented renal function recovery, 13% remained on dialysis after 30 days of follow-up, and 57% died. High-dose continuous PD by flexible catheter and cycler was an effective treatment for ARF. It provided high solute removal, allowing appropriate metabolic and pH control, and adequate dialysis dose and fluid removal. Continuous PD can therefore be considered an alternative to other forms of renal replacement therapy in ARF.