Abstract:

【Abstract】Objective： To compare the predictive ability of the different methods of nutritional assessment on the mortality of elderly hemodialysis patients. Methodology: Sixty patients, who were treated with MHD for more than 3 months at Jun 2010 and aged ≥60 years, were included in this study. SGA, MIS and MNA were used to assess the nutrition in the elderly patients with MHD. They were divided into malnutrition and good nutrition group according to the SGA or MNA-SF and followed up for 36 months. Death or end of the trail was the endpoint of study. Kaplan-Meier and Cox proportional hazard regressions were used to determine the predictive ability of the three methods of nutritional assessment on the mortality of the patients. Results: Kaplan-meier survival analysis showed that the mortality in the malnutrition group was higher than that in the good nutrition group. When grouping was done according to the MIS, there was no different in mortality between malnutrition and good nutrition group. COX regression analysis adjusted for age, kt / v, cardiovascular disease, phosphorus, PTH and calcifacation and other factors, result of SGA was the independent predictor of the death of the patients (OR = 8.877, p = 0.045). But result of MNA-SF had no independent correlation with the death of patients. Conclusion: The predictive ability of SGA on the mortality of the elderly MHD patients is precede of the MIS and the MNA-SF. SGA as overall nutritional assessment method may be more appropriate for elderly MHD patients nutritional assessment.

Key words: maintenance hemodialysis,  nutrition assessment , mortality