In this paper we describe an experimental model designed for studies of continuous ambulatory peritoneal dialysis (CAPD) in uremic rabbits. We preferred the study of uremic animals because it is not known whether peritoneal membrane differs between normal and uremic animals. Animals made uremic after bilateral nephrectomy could not survive on dialysis. Instead, partial nephrectomy of one kidney and partial (5/6) destruction of the cortex of the remaining kidney by electrocauterization provided a simple and reproducible model. CAPD resulted in adequate control of uremia but the animals showed significant decreases in total plasma proteins and weight. This model is suitable for studies of the metabolic complications of CAPD.

We studied the effectiveness, tolerance to, and beneficial metabolic effects of amino acid dialysate over an intermediate period in six CAPD patients. Two liters of 1% amino acid solution (Amino-Dianeal) were alternated with dialysate containing glucose. After four weeks there were significant increases in BUN (from 64 to 102 mg%), total body nitrogen (from 1333 to 1380 g), serum transferrin (from 175 to 222 mg%) and anion gap (from 15.1 to 17.3). Initially, there was a significant rise in HDL cholesterol, however, this was not sustained. No significant change was detected in total-body potassium, fasting serum albumin, triglyceride, insulin, glucagon, electrolytes, anthropometric measurements and daily ingestion of calories and proteins. During the study individual fasting, plasma amino acid levels showed significant increments in respect to histidine, tryptophan and glycine but alanine decreased. Several essential amino acids continued to show values below normal. Two hours after consumption of breakfast and concurrent infusion of the amino acid solution, the plasma levels of the amino acids in the dialysate peaked at emia, which develops in almost onehalf of the CAPD patients (7), and the significant weight gain observed in some of them. Furthermore, the daily losses of albumin and amino acids in the dialysate may induce protein malnutrition, especially if these losses are not replaced by an adequate daily protein intake. The presence of protein malnutrition in CAPD patients is indicated by the low serum albumin and total protein, and by the decrease in total body nitrogen over one year of CAPD (8).

Six non-diabetic CAPD patients were infused over six hours with two litres of a dialysis solution containing 2 g/ dl amino acids (a mixture of essentials and non-essentials). The osmolality of the solution and the amount of ultrafiltration it induced were simiiar to that of a 4.25 g% dextrose Dianeal solution (control), suggesting that an amino acid solution is an efficient osmotic agent. By the end of the six-hour infusion, 80 to 90% of the amino acids present in the dialysis solution had been absorbed. One hour after the infusion was instituted, plasma amino acid levels increased threefold and subsequently decreased to near the initial value by the sixth hour. The amino acid solution was as effective as the dextrose solution in removing urea nitrogen, creatinine and potassium. Our data indicate that intraperitoneal administration of amino acids is effective and well-tolerated in patients on CAPD. We believe further work should be done to determine whether long-term administration of amino acids by this route will improve the nutritional status of these patients and prevent the side effects of daily absorption of large amounts of glucose.