Fat‐free muscle mass in magnetic resonance imaging predicts acute‐on‐chronic liver failure and survival in decompensated cirrhosis

作者：Michael Praktiknjo, Marius Book, Julian Alexander Luetkens, Alessandra Pohlmann, Carsten H. Meyer, Daniel K. Thomas, Christian Jansen, Andreas Feist, Johannes Chang, Jochen M. Grimm, Jennifer Lehmann, Christian Peer Strassburg, Juan G. Abraldeṣ, Guido M. Kukuk, Jonel Trebicka · 发表于：Hepatology · 年份：2017 · DOI：10.1002/hep.29602 · 被引用次数：162 · 研究领域：Nutrition and Health in Aging、Liver Disease and Transplantation、Frailty in Older Adults

Muscle mass seems to be a prognostic marker in patients with liver cirrhosis. However, reported methods to quantify muscle mass are heterogeneous, consented cutoff values are missing, and most studies have used computed tomography. This study evaluated fat-free muscle area (FFMA) as a marker of sarcopenia using magnetic resonance imaging (MRI) in patients with decompensated cirrhosis with transjugular intrahepatic portosystemic shunt (TIPS). The total erector spinae muscle area and the intramuscular fat tissue area were measured and subtracted to calculate the FFMA in 116 patients with cirrhosis by TIPS and MRI. The training cohort of 71 patients compared computed tomography-measured transversal psoas muscle thickness with FFMA. In 15 patients MRI was performed before and after TIPS, and in 12 patients follistatin serum measurements were carried out. The results on FFMA were confirmed in a validation cohort of 45 patients. FFMA correlated with follistatin and transversal psoas muscle thickness and showed slightly better association with survival than transversal psoas muscle thickness. Gender-specific cutoff values for FFMA were determined for sarcopenia. Decompensation (ascites, overt hepatic encephalopathy) persisted after TIPS in the sarcopenia group but resolved in the nonsarcopenia group. Sarcopenic patients showed no clinical improvement after TIPS as well as higher mortality, mainly due to development of acute-on-chronic liver failure. FFMA was an independent predictor...