Introduction: Establishing diagnosis of sarcopenia requires several criteria for assessment which have not been established in CLD. Aim: Determine how well the measures of daily activity level in METs (AAS METs), grip strength normalized by age and gender (GS), and % lean mass (%LM) can be used to assess for secondary sarcopenia in CLD. Methods: Possible secondary sarcopenia (PSS) as defined incorporated pre-sarcopenia (Pre, one criteria indicative of risk), sarcopenia (Sarco, 2 criteria), and severe sarcopenia (Severe, 3 criteria met). The PSS group was compared to subjects with no criteria indicative of sarcopenia (No-PSS). Data from CLD subjects (HCV and NAFLD) and controls across an initial 19 continuous variables pertaining to sarcopenia met all assumptions for inclusion in multivariate analyses. Data included a myokine panel [apelin, fractalkine, erythropoetin, LIF, IL-15, BDNF, irisin, FSTL-1, FABP-3, IL-6, osteonectin, oncostatin, and osteocrin]. Univariate analysis was followed by PSS model analyses using binomial logistic regression. Results: 75 subjects were included (48.0 ± 12.8 years, 62.7% male, 30.1 ± 5.75 BMI, 72.2 ± 9.3 %LM, 8.2 ± 1.3 AAS METs, 3.4 ± 1.6 FSS, 32 possible secondary sarcopenia- 10 presarcopenia, 17 sarcopenia, 5 severe). Comparison of NAFLD, HCV, and Controls showed the expected demographic (age, BMI) and self-report (fatigue and activity) differences. Only serum BDNF was significantly higher Controls versus both HCV and NAFLD (p=<.001). Of patients with NAFLD (N=25), 15 (60%) had PSS (4 Pre/9 Sarco/2 Severe). Of HCV patients (N=30), 13 (43%) had PSS (5 Pre(2 with cirrhosis, C)/5(1C)Sarco/3(1C)Severe). In contrast, of non-liver Controls (N=20), 4 (20%) had PSS (1 Pre/3 Sarco/0)] (only PSS Control vs NAFLD significant at p=0.020). Tests for multicollinearity were negative. Overall model test of fit yielded p<0.001. Omnibus likelihood ratio tests indicate %LM most predicts PSS in the model (p<0.001), followed by GS (p=0.080), and AAS METs (p=0.183). With model predictive cut-off set at 0.5: predictive accuracy 0.853, specificity 0.884, sensitivity 0.813, area under the curve (AUC) 0.938. Absence of PSS was correctly predicted at 88.4%, while presence of PSS was correctly predicted at 81.3%, accurately classifying 64 of 75 subjects. Introduction of myokines individually or in combination did not appear to appreciably improve accuracy, specificity, sensitivity, or AUC. Removal of pre-sarcopenia from cohort analysis significantly worsens model predictive capacity. Conclusions: PSS seems to be higher in NAFLD as compared to controls. %LM is best predictor of PSS as compared to AAS METs and GS. Myokines did not seem to improve the model. Combined, these three easily administered measures can address all three current criteria for sarcopenia and can properly classify 85.3% of subjects regarding presence or absence of PSS.