OBJECTIVES: To evaluate the validity of a leg-to-leg bioimpedance analysis (BIA) system in predicting body composition as measured by dual-energy X-ray absorptiometry (DXA) in postmenopausal women.SUBJECTS AND METHODS: Body fat mass (FM), %Fat and fat free mass (FFM) were measured in 124 postmenopausal women (age: 51–63 y, body mass index (BMI): 17–38 kg/m2) first by the leg-to-leg BIA system, and then by DXA as reference method. Bland-Altman analysis was used to determine the bias and 95% limits of agreement between the two methods for the assessment of the individual. Precision error (CV%) of the BIA system was obtained by repeated measurements with intermediate repositioning.RESULTS: The leg-to-leg BIA system had a high reproducibility with within-day CVs being 0.6% for FFM and 1.1% for FM, and between-day CVs about twice that. The impedance index (Ht2/Z) obtained by the leg-to-leg BIA was moderately correlated to FFM measured by DXA (r=0.66). A significant, systematic bias was observed between the two methods. The BIA system overestimated FM by a mean of 3.1 kg, and underestimated FFM by 2.7 kg. The analysis of 95% limits of agreement showed that for most individuals, %Fat estimated by the BIA might differ from that measured by DXA by 12% below to 45% above, indicating the lack of agreement between the two methods for the assessment of the individual.CONCLUSIONS: The leg-to-leg BIA system can provide simple, rapid and highly reproducible measurements of body composition for groups, but it has limited accuracy for the assessment of the individual. Population-specific equations will be needed to improve its accuracy in estimating body composition in postmenopausal women.