UCSF

Objectives

Frailty is prevalent in patients with cirrhosis and is hypothesized to result in part from sarcopenia, but the precise contribution of sarcopenia to frailty in this population is poorly understood.

Methods

Included were patients with cirrhosis from 2011 to 2014 who had an ambulatory frailty assessment and abdominal computed tomography scan within 3 months. Logistic regression assessed the associations between frailty (=Liver Frailty Index ≥4.5), and sarcopenia (=skeletal muscle index of <39 cm/m for women and <50 cm/m for men).

Results

Two hundred ninety-one participants were included: 33% were female. The median (interquartile range) Liver Frailty Index was 3.7 (3.3-4.2); 19% were frail. The median (interquartile range) skeletal muscle index was 49 cm/m (31-69); 36% had sarcopenia. Among the 54 frail participants, 48% had sarcopenia. In univariable logistic regression, sarcopenia was associated with a 1.86× increased odds of being frail (95% confidence interval [CI], 1.02-3.38). After adjusting for sex, etiology, hepatocellular carcinoma, MELDNa, ascites, encephalopathy, and hypertension, sarcopenia was associated with a 2.38× increased odds of being frail (95% CI, 1.17-4.85). After stratifying by sex and adjusting for MELDNa, sarcopenia among males was associated with a significantly increased odds of frailty (odds ratio 2.81, 95% CI, 1.19-6.67), whereas sarcopenia among females was not (odds ratio 1.38; 95% CI, 0.45-4.25).

Discussion

In patients with cirrhosis, sarcopenia was associated with a nearly 2-fold increased odds of being frail. Two-thirds of frail men displayed sarcopenia compared with only one-quarter of frail women. Contributors to the frail phenotype may differ by sex and support the need for sex-specific strategies to reduce frailty in this population.