Total body dual X-ray absorptiometry is a good measure of both fat mass and fat-free mass in liver cirrhosis compared to "gold-standard" techniques. Melbourne Liver Group.

Liver cirrhosis is a condition in which overnutrition, edema, and undernutrition can coexist simultaneously, or successively, over a period of time, giving rise to alterations in body composition, as well as systemic and multiorgan manifestations. We undertook a cross-sectional study of body composition in 198 adult patients with liver cirrhosis (140 males, mean age 53.6, range 31-85 years; and 58 females, mean age 58.4, range 36-79 years). The patients had cirrhosis of differing etiology and different stages of severity. They were gathered from seven different hospital clinics in the city of Melbourne, Australia, but all the body composition measurements were performed in one body composition laboratory. A variety of body composition techniques were used to identify which commonly available ones could best assess both fat-free mass and fat mass relative to a criterion "gold-standard" method available in a specialist laboratory. A gold-standard fat-free mass (FFMGS) was defined as the sum of total body protein, measured by in vivo neutron activation analysis (IVNA), plus total body water, measured by D2O dilution, plus bone mineral content, measured by dual X-ray absorptiometry (DXA). A gold-standard fat mass (FATGS) was defined as the difference between body weight and FFMGS. "Usual" fat mass and fat-free mass were defined by different techniques including DXA, anthropometry (ANT), single-frequency bioelectrical impedance (SFBIA), multiple-frequency bioelectrical impedance spectroscopy (MFBIA), and whole body gamma counting (TBK). The FFMGS was overhydrated in both sexes, relative to the usual value of 0.73, but women were significantly overhydrated compared to men. Relative to the gold-standard deuterium oxide dilution method for measuring total body water, SFBIA slightly overestimated TBW, whereas MFBIA slightly underestimated TBW, with both methods having wide limits of agreement for any single estimate. In comparing FFM to FFMGS, only DXA showed a small negative bias, in both males and females, with modest limits of agreement for any single estimate. All other methods showed a large negative bias (ANT, SFBIA, and MFBIA) or a large positive bias (TBK) relative to FFMGS, with wide limits of agreement. In comparing FAT with the FATGS, only DXA showed a small positive bias, in both males and females, with modest limits of agreement for any single estimate. All other methods showed a large positive bias (ANT, SFBIA, and MFBIA) or a large negative bias (TBK) relative to FATGS, with wide limits of agreement. In cirrhosis, DXA is a good and widely available method to assess both fat mass and fat-free mass. However, it cannot give information about the quality of the FFM, particularly its water content. The bedside methods of anthropometry and bioelectrical impedance, both SFBIA and MFBIA, are poor methods of measuring body composition in patients with liver cirrhosis, whereas whole body gamma counting, although not widely available, also significantly differs from the gold-standard method of assessment of fat-free mass and fat mass in liver cirrhosis.