Evaluation of techniques used to assess skeletal muscle quantity in patients with cirrhosis.

Abstract

Loss of skeletal muscle mass is a well-recognised complication of cirrhosis. Bedside methods to assess skeletal muscle mass including anthropometrics and bioelectrical impedance analysis (BIA) are negatively impacted by fluid overload in advanced cirrhosis and thus there is a need to identify alternatives. There is a paucity of data on the accuracy of commonly used radiological methods such as dual X-ray absorptiometry (DXA) to assess appendicular lean mass (ALM), and computed tomography (CT) skeletal muscle area in patients with cirrhosis. The aim of this study was to evaluate the relationships and agreement of several skeletal muscle mass estimation methods compared to a reference model in patients with cirrhosis. A cross-sectional, single centre study was performed by prospectively recruiting patients with cirrhosis referred to the Queensland Liver Transplant Service. Patients underwent assessment of skeletal muscle mass using bedside techniques (mid-upper arm muscle circumference (MUAMC), bioelectrical impedance spectroscopy (BIS), ultrasound muscle thickness (USMT)) and radiological methods (DXA ALM, CT skeletal muscle area). These were compared to a reference measurement of body cell mass derived from a multi-compartment model using isotope dilution tests and DXA. Forty-two patients (age 56 years, interquartile range 48-60, 86% male) were recruited. Bedside skeletal muscle mass estimation techniques were strongly correlated to the body cell mass reference, with BIS estimation having the strongest correlation coefficients (r=0.78-0.79; P<0.01). A novel technique measuring USMT offered no advantage over traditional bedside techniques. Of the radiological methods, DXA ALM had the strongest correlation coefficient (r=0.781; P<0.01). Weaker correlation coefficients were observed in patients with ascites, except when using the MUAMC. Bland-Altman analysis of BIS body composition estimates demonstrated significant systematic biases and large limits of agreement compared to reference values. These results confirm the difficulties in assessing skeletal muscle mass in patients with cirrhosis, particularly in those with ascites. DXA ALM and BIS measurements provided the best correlation to body cell mass. We suggest DXA ALM for estimation of skeletal muscle mass in patients with cirrhosis as there are established thresholds for skeletal muscle mass depletion, and an accurate assessment of bone mass and density can also be provided. The use of USMT over other bedside skeletal muscle mass estimates was not supported by our results. Further studies evaluating novel bedside skeletal muscle mass estimation techniques in cirrhosis patients are required.