RESUMO
OBJECTIVES: Nutritional diagnosis is essential given that malnutrition negatively affects morbidity and mortality in patients with liver disease. In the absence of a standard method, limited accuracy has been reported in relation to nutritional assessment. The aim of the study was to evaluate the use of subjective global assessment (SGA) and different methods of nutritional assessment, isolated and in combination with SGA to predict clinical outcomes. METHODS: This was a longitudinal study with patients waiting for liver transplantation. Nutritional status was classified according to SGA. Anthropometric parameters, standard phase angle (SPA), handgrip strength, and 6-min walk test (6MWT) were evaluated. Univariate and multivariate analysis and receiver operator characteristic curve were performed. P < 0.05 was statistically significant. RESULTS: Seventy-three patients with an average age of 52.3 ± 11.4 y were evaluated. Of these, 63% were men. Low midarm muscle area (MAMA; P < 0.001), malnutrition by SGA + low MAMA (P < 0.001), and malnutrition by SGA + low SPA (P < 0.001) were independent predictors of clinical features of advanced cirrhosis. SGA + SPA presented the greater area under the curve (AUC: 0.6431). Death was predicted by low midarm circumference (MAC; P = 0.037) and slow gait speed on the 6MWT (GS-6MWT; P = 0.017), with both parameters isolated or concomitantly linked to malnutrition. SGA + GS-6MWT presented the highest predictive power for death (AUC: 0.6673) slightly greater than SGA+MAC (AUC: 0.6346). CONCLUSION: The results indicate that SGA together with SPA has greater predictive power for features of advanced cirrhosis, and SGA plus gait speed is able to better predict mortality in patients on the waiting list for liver transplant.
