Abstract: 【Abstract】Objective Nutritional status can be evaluated by serum albumin, modified quantitative subjective global assessment (MQSGA) and malnutrition-inflammation score (MIS). This study sought to determine the value of these methods for malnutrition assessment in peritoneal dialysis (PD) patients. Methods
Fifty-seven clinically stable patients undergoing PD were recruited for the evaluation of nutritional status by serum albumin, MQSGA and MIS. Their anthropometry and biochemical assays were evaluated. Correlation among the results from the 3 methods was conducted. Receiver operating characteristic (ROC) curve was applied to study the sensitivity and specificity of the 3 methods for the evaluation of malnutrition status. Logistic proportional hazard regression model was used to assess the risks during hospitalization. Results ①Malnutrition status was found in all PD patients by MIS, in 70.2% patients by MQSGA, and 57.9% patients by serum albumin. ②MQSGA and MIS correlated to anthropometry, biochemical assays and hospitalization frequency. Nutritional assessments by MQSGA, MIS and serum albumin were significantly correlated (P＜0.01). ③Area under the curve of MQSGA, MIS and serum albumin was 0.966, 0.961 and 0.028, respectively, for malnutrition diagnosed by PEW. ④Logistic proportional hazard regression demonstrated that MQSGA score had the highest correlation with the risks during hospitalization (RR 2.559, CI 1.498～4.372, P=0.001). Conclusions MIS is a sensitive method for the evaluation of malnutrition in PD patients. MQSGA is similar to MIS for malnutritional assessment in PD patients. MQSGA may be a better predictor for risks during hospitalization in PD patients.

Key words: peritoneal dialysis, malnutrition, modified quantitative subjective global assessment（MQSGA）, malnutrition-inflammation score（MIS）