Assessment of Body Composition of Rats by Bioimpedance Spectroscopy: Validation Against Dual-Energy X-Ray Absorptiometry

Abstract

Fat-free mass (FFM) and fat mass (FM) were determined in male and female out-bred Wistar strain rats by  both dual-energy X-ray absorptiometry (DXA) and by bioimpedance spectroscopy (BIS). Data obtained by  both methods were highly correlated and exhibited a small (4%) bias for FFM, with relatively small limits  of agreement (approximately ± 10%), but differed to a much larger degree for FM (14 to 27% bias), with  wide limits of agreement (± 50%). Inter-method correction equations are provided to allow conversion of  data from one method to another, eliminating bias, but not altering the limits of agreement. Since both  methods exhibited a high degree of precision of measurement, it is suggested that the poor agreement for  measurement of fat mass is a reflection that fat mass is an indirectly-derived value, which includes unavoidable  propagation of prediction errors associated with the primary measures. Both BIS and DXA provide  rapid, minimally invasive and, in the case of BIS, portable techniques for body composition analysis. Their  use for estimation of FFM may be recommended, but, for prediction of fat mass, exercise of caution would  be prudent.