To test the validity of the determination of protein catabolic rate (PCR) as a measure of protein intake in dialysis patients, we studied a selected population of 27 well-dialysed patients, free of catabolic illnesses, and in apparent clinical stability. Daily protein intake, obtained by controlled dietary records, and protein catabolic rate, measured by urea kinetic modeling, were 1.02 +/- 0.26 and 0.99 +/- 0.16 g/kg respectively. Although there was a good correspondence between protein intake and protein catabolic rate in many of our patients, in the six cases with dietary protein content less than 0.8 g/kg, protein catabolic rate was regularly greater than 0.8 g/kg; on the contrary, in four patients ingesting more than 1.2 g/kg of protein, protein catabolic rate values were significantly less than protein intake. Our results demonstrate that some degree of nitrogen imbalance may be present in about 25% of clinically stable dialysis patients, and suggest that current concepts of clinical stability do not imply a stable metabolic state in a substantial portion of dialysis patients. Therefore, protein catabolic rate determination does not provide a reliable index of protein intake in many dialysis patients.