OBJECTIVE: To assess the application of a Cole-Cole analysis of multiple frequency bioelectrical impedance analysis (MFBIA) measurements to predict total body water (TBW) and extracellular water (ECW) in humans. This technique has previously been shown to produce accurate and reliable estimates in both normal and abnormal animals. DESIGN: The whole body impedance of 60 healthy humans was measured at 496 frequencies (ranging from 4 kHz to 1 MHz) and the impedance at zero frequency, Ro, and at the characteristic frequency, Zc, were determined from the impedance spectrum, (Cole-Cole plot). TBW and ECW were independently determined using deuterium and bromide tracer dilution techniques. SETTING: At the Dunn Clinical Nutrition Centre and The Department of Biochemistry, University of Queensland. SUBJECTS: 60 healthy adult volunteers (27 men and 33 women, aged 18-45 years). RESULTS: The results presented suggest that the swept frequency bioimpedance technique estimates total body water, (SEE = 5.2%), and extracellular water, (SEE = 10%), only slightly better in normal, healthy subjects than a method based on single frequency bioimpedance or anthropometric estimates based on weight, height and gender. CONCLUSIONS: This study has undertaken the most extensive analysis to date of relationships between TBW (and ECW) and individual impedances obtained at different frequencies ( > 400 frequencies), and has shown marginal advantages of using one frequency over another, even if values predicted from theoretical bioimpedance models are used in the estimations. However in situations where there are disturbances of fluid distribution, values predicted from the Cole-Cole analysis of swept frequency bioimpedance measurements could prove to be more useful.

Type

Journal article

Journal

Eur J Clin Nutr

Publication Date

03/1996

Volume

50

Pages

159 - 164

Keywords

Adolescent, Adult, Anthropometry, Body Composition, Body Mass Index, Body Water, Electric Impedance, Evaluation Studies as Topic, Extracellular Space, Female, Humans, Male, Middle Aged, Regression Analysis, Sex Factors