Derivation and validation of a simple anthropometric equation to predict fat-free mass in patients with chronic hepatitis C.

BACKGROUND: Loss of skeletal muscle mass is very common in chronic liver diseases and affects 30.0-70.0% of the patients with cirrhosis. Given the relevance of muscle wasting in hepatic diseases, a practical screening tool for earlier detection of skeletal muscle mass loss is of utmost significance. AIMS: To develop and validate a simple anthropometric prediction equation for fat-free mass estimation by using Bioelectrical Impedance Analysis (BIA) as a reference method in patients with chronic hepatitis C (CHC). METHODS: We prospectively, included 209 CHC patients, randomly allocated into two groups, 158 patients in a development model (derivation sample) and 51 patients in a validation group (validation sample). Predictive equations were developed using backward stepwise multiple regression and the most adequate and simplest derived predictive equation was further explored for agreement and bias in the validation sample. The accuracy of the predictive equation was evaluated using the coefficient of determination (R2). RESULTS: The predictive equation with an optimal R2 was Fat-Free Mass (Kg) = Sex × 0.17 + Height (m) × 16.83 + Weight (Kg) × 0.62 + Waist Circumference (cm) × (-0.15) + Weight (Kg) × Sex × (-0.30) + Sex × Waist Circumference (cm) × 0.14-6.23; where sex = 1 for female and 0 for male. R2 = 0.93, standard error of the estimate = 2.6 Kg and coefficient of variation = 20.0%, p < 0.001. CONCLUSIONS: Our developed and cross-validated anthropometric prediction equation for fat-free mass estimation by using BIA attained a high coefficient of determination, a low standard error of the estimate, and lowermost coefficient of variation. This study indicates that predictive equations may be reliable and useful alternative methods for clinical evaluation of fat-free mass in patients with CHC.