Many epidemiologic studies indicate that protein-energy malnutrition is a strong predictor of morbidity and mortality in maintenance hemodialysis and chronic peritoneal dialysis patients. Those parameters of protein-energy malnutrition that have been most clearly associated with morbidity or mortality include serum albumin, dietary protein intake as indicated by the protein equivalent of total nitrogen appearance (PNA, also referred to as PCR), and the predialysis serum urea nitrogen. In large scale clinical surveys, low predialysis serum creatinine, cholesterol, potassium, phosphorus, calcium, and bicarbonate also correlate with increased mortality rates in maintenance hemodialysis patients. These correlations may reflect an association between dietary intake and mortality. The paucity of data correlating energy intake or body composition (e.g., total body protein or nitrogen, skeletal muscle mass, total body fat) with clinical outcome may reflect the difficulty of obtaining these data in large scale prospective or retrospective analyses. The correlation between protein-energy malnutrition and increased morbidity and mortality rates does not prove that a higher protein and/or energy intake or a more optimal body composition will improve the patient's outcome. Prospective, randomized studies will be necessary to examine this question. However, two retrospective analyses of the effect of intradialytic parenteral nutrition on mortality rates in maintenance hemodialysis patients are consistent with the thesis that increased nutrient intake will reduce mortality in maintenance hemodialysis patients. Methods for assessing protein-energy nutritional status are discussed. There are essentially no data indicating that the more complicated and expensive techniques for nutritional assessment have important advantages over the more simple methods for the clinical management of maintenance dialysis patients.