Comparing the predictive ability of population-specific Mini-Nutritional Assessment with Subjective Global Assessment for Taiwanese patients with hemodialysis: a cross-sectional study.

BACKGROUND: The Mini-Nutritional Assessment (MNA) and Subjective Global Assessment (SGA) are two frequently used tools in nutritional assessment. OBJECTIVES: This study aimed to evaluate the feasibility of applying the MNA with population-specific anthropometric modifications and to compare the predictive ability of MNA with that of the SGA in patients with hemodialysis. DESIGN: Purposive sampling. METHODS: This study was conducted in the Hemodialysis Unit of E-Da Hospital in Kaohsiung, Taiwan. A total of 192 patients with hemodialysis were evaluated their nutritional status concomitantly with the SGA and the MNA in two versions-MNA Taiwan version-I adopted population-specific anthropometric cut points, and MNA Taiwan version-II had body mass index (BMI) omitted. RESULTS: The SGA graded 1% malnourished and 51% at risk of malnutrition for patients with hemodialysis; MNA Taiwan version-I graded 3% and 41%, respectively, whereas MNA Taiwan version-II graded 5% and 36%, respectively. There were significant differences between patterns of nutritional status predicted with the SGA and the MNA versions (P<0.05). The scores predicted with both tools correlated positively with appetite, serum albumin and creatinine levels, BMI, and mid-arm and calf circumferences, and negatively with number of emergency visits. However, only MNA versions negatively correlated with length of hospital stay (all P<0.05). Linear regression analysis revealed that the MNA Taiwan version-I, -II and SGA scores were positively associated with BMI and serum albumin level, and negatively associated with number of emergency visits after adjusting for confounders (all P<0.05). However, only MNA Taiwan version-I and -II were negatively associated with C-reactive protein (CRP) level (P<0.05). The associations of MNA Taiwan version-II with albumin and CRP levels were stronger than those obtained using MNA Taiwan version-I and SGA. CONCLUSIONS: The present study suggests that both the modified MNA versions can evaluate nutritional risk of patients with hemodialysis in Taiwan. The MNA Taiwan version-II which adopted population-specific anthropometric cut values without BMI is better able to assess nutritional status and reflect health status of patients with hemodialysis than MNA Taiwan version-I and SGA.