An improved method to assess skeletal muscle mass in patients with liver cirrhosis based on computed tomography images

Aim Conventionally, the skeletal muscle area with computed tomography (CT) attenuation ranging from -29 to +150 Hounsfield unit (HU) divided by height squared (the conventional skeletal muscle index [SMI]) was used as an index of skeletal muscle mass. However, it includes fat-infiltrated skeletal muscle, which is known to have poor function. This study aims to determine whether the low-fat SMI, which uses skeletal muscle mass with CT attenuation ranging from +30 to +150 HU, or conventional SMI appropriately reflects the function of skeletal muscle. Methods We retrospectively analyzed 120 patients with cirrhosis whose handgrip strength was measured. Among them, 48 patients underwent a physical performance assessment such as liver frailty index (LFI) and short physical performance battery (SPPB), and 80 underwent quality of life (QOL) assessment. The relationships between each SMI and handgrip strength, LFI, SPPB, and QOL were evaluated. Results Low-fat SMI was significantly correlated with handgrip strength (males, R = 0.393, p = 0.002; females, R = 0.423, p < 0.001) and LFI (males, R = -0.535, p = 0.035; females, R = -0.368, p = 0.039), whereas conventional SMI was not. When using low-fat SMI, patients with low skeletal muscle mass had significantly low handgrip strength, LFI, SPPB, and physical and social-related QOL score than those without. By contrast, no significant differences were found for any items when using conventional SMI. Conclusions Low-fat SMI is a good index of skeletal muscle mass that appropriately reflects skeletal muscle function.