Frequency and Clinical Impact of Sarcopenia in Cirrhotic Patients with and without Hepatocellular Carcinoma

Abstract

Purpose: Prognostic assessment of patients (pts) with cirrhosis remains a challenge, as the natural history is highly variable due to several factors, including hepatic synthetic function, and development of hepatocellular carcinoma (HCC) among others. Sarcopenia is defined as low levels of muscle mass, and this syndrome may be present in chronic and malignant diseases; however, it is not well studied and understood in pts with cirrhosis and pts with HCC. We aimed to establish the frequency of sarcopenia and if it predicts mortality in a cohort of cirrhotic pts with and without HCC. Methods: 163 pts with cirrhosis that were consecutively evaluated for liver transplant and had a computed tomography (CT) scan at the 3rd lumbar vertebrae were selected. Data were recovered from medical charts, skeletal muscle cross-sectional area was measured by CT in order to determine the L3 skeletal muscle index (L3 SMI) defined as total lumbar muscle cross sectional area adjusted for stature, and sarcopenia was defined using previously published gender-specific cutoffs. Results: 118 pts were males (72%), the mean age was 55±1 years, and 51 pts (31%) had HCC at the time of the CT. Median time of follow-up was 19±1 months. Etiology of liver cirrhosis was HCV (34%), alcohol (20%), autoimmune liver disease (13%), HBV (6%), and others (27%). Sarcopenia was present in 61 pts (37%), and there was no difference in the frequency of sarcopenia among pts with and without HCC (31% vs. 40%, P=0.3). By univariate Cox analysis the bilirubin (HR 1.01; P<0.001), INR (HR 6.42; P<0.001), creatinine (HR 1.01, P=0.001), albumin (HR 0.95, P=0.006), MELD (HR 1.12; P<0.001), Child-Pugh (HR 2.09; P<0.001), sodium (HR 0.89; P<0.001), L3 SMI (HR 0.98, P=0.05), and sarcopenia (HR 2.19; P=0.001) were associated with mortality. The presence of HCC was not associated with increased mortality (HR 1.12; P=0.6). By multivariate Cox regression analysis only the MELD score (HR 1.08, P=0.001), and sarcopenia (HR 2.18, P=0.001) were independently associated with early mortality. Median survival for sarcopenic pts was 19±5 months, compared to 30±2 months in non-sarcopenic pts (P=0.001). There was a poor correlation between sarcopenia and MELD score (r -0.13, P=0.1) and sarcopenia and Child-Pugh score (r 0.04, P=0.6). Conclusion: Sarcopenia is present in more than one third of pts with cirrhosis, and is not associated with the presence of HCC. Sarcopenia constitutes a strong and independent predictor of mortality in pts with cirrhosis and does not correlate with degree of liver dysfunction evaluated with conventional scoring systems. Further studies that include sarcopenia with conventional scores may allow significantly better prediction of mortality among cirrhotic pts with and without HCC.