Nutritional status and survival in end-stage renal disease patients.

Abstract

Several reports have emphasized that putative laboratory surrogates of nutrition, such as serum albumin, creatinine, and cholesterol concentrations are statistically more powerful independent predictors of odds risk of death for dialysis patients than is the delivered dose of dialysis. In view of the relative simplicity with which these blood tests can be obtained, their lack of expense, and simplicity in interpretation, the dialysis community has greatly escalated their importance as performance measures for the processes of patient care, arguably without full consideration of their meaning. If malnutrition in dialysis patients is a powerful predictor of death risk, and is amenable to corrective interventions that result in a reduction in the odds risk of death, then the zeal with which these laboratory tests have been embraced is appropriate. However, the assumption that a statistical correlation between laboratory surrogates of malnutrition, or other measures of inadequate nutrition, such as body mass index or a subjective global assessment, indicate a direct causal relationship between nutritional intake, nutritional status, and outcome may be incorrect. Such apparent linkages may be a consequence of the statistical model selected alone, i.e., another unappreciated medical condition may be the proximate cause of death in addition to resulting in malnutrition. The mechanism(s) by which malnutrition may adversely impact the survival of end-stage renal disease (ESRD) patients is unclear. The impact of milder degrees of malnutrition on patient survival, their proximate effect on survival, and the reality of their independent effect on patient survival are also inadequately defined. Clearly, there is a statistical link between the putative laboratory surrogates of nutrition and patient survival. Regardless of the pathobiology of such a causal link, it is valid to enquire if an intervention that results in a positive change in nutritional parameters enhances patient survival. These issues surrounding nutritional status and survival in patients with ESRD are reviewed here in detail. The conclusion of this critique is that additional studies are needed to determine if malnutrition is truly an independent and responsive predictor of outcome for ESRD patients.