Comparison of bipolar and tetrapolar impedance techniques for assessing fat mass

Whole body fat mass (BF) can be predicted by different bioimpedance (BIA) techniques measuring the whole body (tetrapolar hand-to-foot technique, HF), lower body (bipolar foot-to-foot technique, FF), or upper body (bipolar hand-to-hand technique, HH). This study analyzed 146 healthy volunteers (age 18-84 years) for whether these three techniques differ in their estimates of BF, and whether sex and age of the subjects influence estimates of BF. Reproducibility of the techniques was tested by calculating the technical error (TE). Effects of BIA technique, sex, and age on predicted BF were analyzed with ANOVA for repeated measurements in a mixed effect design. Results showed high reproducibility for the three BIA techniques (TE = 0.01-0.03 kg). ANOVA indicated interactions between BIA technique and sex (P = 0.035), BIA technique and age (P < 0.001), as well as effects of sex (P = 0.004) and age (P = 0.001) on variation in BF. The HH technique gave the highest values for BF in males, but lowest values in females, whereas the reverse was found for the FF technique. The HH technique yielded the lowest values for BF in young adults, but highest values in older ones. The reverse was noted for the FF technique. The data suggest that the observed differences in the three BIA techniques in predicting BF reflect sex differences and age-associated changes in body fat patterning. Therefore, the whole body impedance method is preferred over the HH and FF techniques due to the interactions with sex and age. (C) 2004 Wiley-Liss, Inc.