Urea kinetic modelling (UKM) was used to assess adequacy of dialysis in 50 CAPD patients. Nutritional status was assessed from the measurement of visceral protein status (total protein, albumin, transferrin, immunoglobulins, complement), somatic protein status (anthropometry), and dietary intake (1 week weighed dietary inventory and normalized protein catabolic rate (NPCR) from UKM). Morbidity was assessed from the peritonitis and admission history. Mean Kt/V (corrected to x3 weekly dialysis) was 0.66 +/- 0.02. Dietary protein intake estimated from the NPCR (1.08 +/- 0.03 g kg-1 day-1) correlated well (r = 0.72, P < 0.001) with that estimated from the dietary inventory (1.10 +/- 0.04 g kg-1 day-1). There was a strong correlation between Kt/V and NPCR corrected for actual weight (r = 0.65, P < 0.001), but when NPCR was corrected for IBW this correlation was weaker (r = 0.35, P < 0.05). Patients were divided by Kt/V into two groups (> 0.65, n = 22 and < 0.65, n = 28). There were no significant differences in the indices of visceral protein status between the two groups. Weight, height, BMI, fat free mass and arm muscle area were significantly greater in the group Kt/V < 0.65. Residual renal function (creatinine clearance) was higher in the group Kt/V > 0.65 (3.8 +/- 0.7 versus 1.9 +/- 0.5 1/24 h, P < 0.05) and plasma creatinine less (913 +/- 51 versus 1265 +/- 51 mumol/l, P < 0.001). Hb, potassium, bicarbonate, phosphate, alkaline phosphatase, PTH, and blood pressure were not different. Neither was there any difference between the two groups in any of the indices of morbidity.