Combined evaluation of nutrition and hydration in dialysis patients with bioelectrical impedance vector analysis (BIVA).

Abstract BACKGROUND & AIMS: Body hydration changes continuously in hemodialysis patients. The Subjective Global Assessment (SGA) is used for the nutritional evaluation but it does not allow a direct evaluation of hydration. Bioelectrical impedance vector analysis (BIVA) is very sensitive to hydration. The potential of the combined evaluation of hydration and nutrition with SGA and BIVA is still lacking. METHODS: Observational cross-sectional study on 130 (94 Male) uremic patients undergoing chronic hemodialysis three times a week. Nutritional status was evaluated with the SGA. Each subject was classified as SGA-A (normal nutritional status), SGA-B (moderate malnutrition), or SGA-C (severe malnutrition). Body hydration was evaluated with BIVA. The two vector components resistance (R) and reactance (Xc) were normalized by the subject's height and standardized as bivariate Z-score, i.e. Z(R) and Z(Xc). RESULTS: Undernutrition influenced impedance vector distribution both before and after a dialysis session. In pre-dialysis, the mean vector of SGA A was inside the 50% tolerance ellipse. In SGA B and C, Z(R) was increased and Z(Xc) decreased, indicating a progressive loss of soft tissue mass. Fluid removal with dialysis increased both Z(R) and Z(Xc) in SGA A and B but not in C. With ROC curve analysis on the slope of increase, we found the cutoff value of 27.8° below which undernutrition was present, either moderate or severe. The area under the ROC curve was 77.7° (95% CI 69.5-84.5, P < .0001) with sensitivity 75.9%, specificity 78.6%, positive predicted value 74.6%, and negative predicted value 79%. CONCLUSIONS: The distribution of impedance vectors is associated with the SGA classification of patients. The change in body hydration in each SGA category can be detected with BIVA.