Comparison of nutritional risk screening tools for predicting sarcopenia in hospitalized patients

Background/aim: The aim of this study was to assess the risk of malnutrition in hospitalized patients with three different tests and to compare these tests in terms of long hospitalization periods and sarcopenia. Materials and methods: Hospitalized patients in an internal medicine clinic were enrolled in this cross-sectional study. Patients were grouped as under 65 years (Group 1 = G1) and over 65 years old (Group 2 = G2). The nutritional status of the patients was evaluated with the Nutritional Risk Screening (NRS) 2002, Universal Malnutrition Screening Tool (MUST), Mini Nutritional Assessment Short Form (MNA-SF), and total Mini Nutritional Assessment (MNA) tests. Diagnosis of sarcopenia was assessed via bioimpedance analysis for muscle mass, a hand-grip strength test, and a "timed get up and go" test. Nutritional tests were compared in terms of sarcopenia and long hospitalization periods with receiver operating characteristic curve analysis. Results: Mean ages were 54 (G1, n = 84) and 76 (G2, n = 112) years old. Sarcopenia was found in 5% in G1 and 33% in G2. The MNA-SF in G1 (area under curve (AUC) = 0.585, P = 0.26; sensitivity 41%, specificity 44%) and the MUST in G2 (AUC = 0.614, P = 0.048; 25%, 86%) were better predictors of prolonged hospitalization. The MNA-SF was associated with sarcopenia in both groups (G1: AUC = 0.716, P = 0.147; 63%, 64% and G2: AUC = 0.762, P < 0.001; 86%, 48%). In addition, the MNA-SF was a better predictor of low lean muscle mass index (AUC = 0.762, P < 0.001; 86%, 48%), low grip strength (AUC = 0.594, P = 0.27; 65%, 50%), and reduced walking speed (AUC = 0.642, P = 0.01; 71%, 47%) in G2. Conclusion: None of the three tests are highly sensitive or specific for predicting sarcopenia. The MNA-SF is a better test to evaluate sarcopenia and/or related parameters than the others, and the MUST is related to prolonged hospitalization in older patients.