Abstract
Diagnosing sarcopenia is challenging. This multicenter cross-sectional study aimed to evaluate the utility of the SARC-F score system for identifying sarcopenia in patients with chronic liver disease (CLD). We enrolled 717 patients from five participating centers who completed the SARC-F between November 2019 and March 2021. Sarcopenia was diagnosed based on the Japan Society of Hepatology Working Group on Sarcopenia in Liver Disease Consensus. Muscle strength was estimated using a grip dynamometer, and muscle mass was assessed using computed tomography or bioelectrical impedance analysis. The association between SARC-F and sarcopenia was analyzed using a logistic regression model. The optimal SARC-F cutoff value for identifying sarcopenia was determined using receiver operating characteristic (ROC) curve analysis. Of the 676 eligible patients, 15% were diagnosed with sarcopenia. The SARC-F distribution was 0 points in 63% of patients, 1 point in 17%, 2 points in 7%, 3 points in 4%, and ≥4 points in 8%. The SARC-F items of "Strength" (odds ratio (OR), 1.98; 95% confidence interval (CI), 1.03-3.80) and "Falls" (OR, 2.44; 95% CI, 1.48-4.03) were significantly associated with sarcopenia. The SARC-F value of 1 point showed a higher discriminative ability for identifying sarcopenia than the 4 points that are conventionally used (p < 0.001), with an area under the ROC curve of 0.68, sensitivity of 0.65, specificity of 0.68, positive predictive value of 0.27, and negative predictive value of 0.92. SARC-F is useful for identifying patients with CLD who are at risk of sarcopenia.