Energy malnutrition worsens survival in patients with liver cirrhosis, and is currently defined as non-protein respiratory quotient (npRQ) <0.85, as measured by indirect calorimetry. However, measurement of this npRQ is limited because of the high cost of indirect calorimetry. Therefore, we sought an alternative marker that can be used in the routine clinical setting. Forty-four inpatients with cirrhosis were recruited in this study. The last meal was served at 18:00 h on the previous day, and indirect calorimetry was performed between 07:00 and 09:00 h while the patients were still in bed. Fasting blood samples were collected in the early morning on the day of the test. Anthropometry was performed by an expert dietician. The correlations among npRQ, Child-Pugh score of disease severity, laboratory parameters, %AC (arm circumference), %TSF (triceps skinfold thickness), and %AMC (arm muscle circumference) were studied using simple linear regression analysis. ROC (Receiver operating characteristic) analysis was used to identify the cut-off values that would best predict npRQ=0.85. npRQ correlated significantly with %AC (r(2)=0.204, p=0.0021) and %AMC (r(2)=0.178, p=0.0043) but not with %TSF. npRQ was not significantly correlated with other laboratory or anthropometric measurements. The cut-off value for %AC that showed the largest AUC (area under the curve) by ROC analysis was 95, while that for %AMC was 92. Multiple regression analysis yielded an equation; npRQ=0.0019×(%AC)20.0134×(Child-Pugh score)+0.7791. Patient stratification by %AC=95 or by regression equation-based npRQ=0.85, but not by %AMC=92, produced significant difference in survival curves. %AC and regression equation could represent npRQ to some extent as parameters of energy nutrition in cirrhosis.