During treatment of patients with non-insulin-dependent diabetes mellitus,
there may be marked body weight loss. Therefore, body composition should be
monitored to check for a decrease in fat mass alone, without an excessive
decrease of both fat-free mass and total body water. Accordingly, it is use
ful to monitor the hydration of these patients. One method that allows us t
o check the status of body hydration is the multifrequency bioelectric impe
dance analysis (MFBIA). It makes use of formulas that estimate total body w
ater on the basis of the concept that the human body may be approximated to
a cylinder of length equal to body height. In normal subjects body water e
stimates are sufficiently accurate, but in obese subjects the true hydratio
n status may be overestimated. In this report, we describe the accuracy of
mathematical models previously described in the literature, and correct for
the overestimation of total body water in obese subjects by means of a new
equation based on a new model. The coefficients for each model have been r
ecalculated by the weighing of our sample in order to test the accuracy of
estimates obtained with the equations. This new model includes both body vo
lume and two impedances at appropriate frequencies useful for identifying t
wo terms strictly related to extra- and intra-cellular water. The new formu
las do not include body weight, but they include the body volume, a paramet
er more closely related to the biophysical reference model. Fifty-five over
weight females, body mass index ranging from 26.8 to 50.2 kg/m(2), were enr
olled in the study. The proposed equations, taking advantage of two impedan
ce values at appropriate frequencies, better predict total body water in ob
ese women. This was particularly evident when the results obtained with the
multifrequency bioelectric impedance analysis and deuterium isotopic oxide
dilution method were compared. Although this last method is considered the
"gold standard," it is not suitable for use in routine clinical practice.
In conclusion, evaluation of total body composition by means of bioelectric
impedance analysis might be included in programs for the prevention of non
-insulin-dependent diabetes and for monitoring weight loss during overt pat
hology.