A prospective randomised trial of intraperitoneal sodium nitroprusside (SNP) administration on the efficacy of acute intermittent peritoneal dialysis was carried out in 40 adult patients of acute or acute on chronic renal failure. A total of 36 cycles of peritoneal dialysis (PD) with an exchange volume of 1 litre and duration of 1 hour per cycle were given to each patient. The 36 cycles of PD were divided into 12 clearance periods (I-XII) of 3 cycles each. SNP was added in a dosage of 4 mg/l of dialysate in clearance period II, IV, VI and VIII. Of 40 patients, 20 were subjected to standard PD (Gp A) while the other 20 received SNP added PD (Gp B). The peritoneal clearance of urea, creatinine, percentage fall of blood urea, serum creatinine and protein loss during the various clearance periods were compared in the two groups. It was observed that group B patients had significantly higher peritoneal clearance and the percentage reduction in the blood levels of urea and creatinine. Protein loss per clearance period was also significantly higher in group B patients. Twenty two cycles of SNP added PD were as effective as 36 cycles of standard PD. No systematic untoward effects of SNP were observed. It is therefore, concluded that intraperitoneal SNP administration is a safe and effective way of increasing the efficacy of PD thus reducing the duration of treatment.