Abstract
This study aimed to evaluate whether the subjective global assessment (SGA) could effectively predict energy malnutrition, as assessed by indirect calorimetry, and mortality in hospitalized patients with cirrhosis. Energy malnutrition was defined by a nonprotein respiratory quotient (npRQ) < 0.85 using an indirect calorimetry. The usefulness of the SGA in identifying energy malnutrition and predicting mortality was assessed by the logistic regression and Cox proportional hazards models, respectively. Out of the 230 patients analyzed, 43% were found to have energy malnutrition. The distribution of SGA classifications was 54% for SGA-A, 32% for SGA-B, and 14% for SGA-C. Multivariable analysis indicated that both SGA-B (odds ratio, 3.59; 95% confidence interval [CI], 1.59-8.10) and SGA-C (odds ratio, 19.70; 95% CI, 3.46-112.00), along with free fatty acids (FFA), were independently linked to energy malnutrition. Regarding mortality, 125 patients (54%) died over a median follow-up period of 2.8 years. After adjustment, SGA-B (hazard ratio, 1.81; 95% CI, 1.08-3.03) and SGA-C (hazard ratio, 3.35; 95% CI, 1.28-8.76) were predictors of mortality in cirrhosis patients, while energy malnutrition and FFA were not. The SGA is a valuable tool for identifying energy malnutrition and predicting mortality in patients with cirrhosis.