Validity of total body water measured by multi-frequency bioelectrical impedance devices in a multi-ethnic sample

ABSTRACT
Background & aims: Total body water (TBW) is the largest component of fat free mass and therefore is commonly used in multi-compartment body composition models and as a stand-alone method to assess body composition. Previous literature has not validated bioelectrical impedance spectroscopy device estimates of TBW in racial and ethnic minority populations; previous studies have focused on bioelectrical impedance analysis devices that rely on proprietary algorithms and participant characteristics for accuracy. The purpose of this study was to assess the validity of two bioelectrical impedance spectroscopy devices for measures of TBW compared to a deuterium dilution criterion in a multi-ethnic sample.
Methods:
109 individuals (55% female, Age: 26.6 ± 6.9 yrs, BMI: 25.3 ± 4.0 kg/m2) identifying as Asian, African American/Black, Caucasian/White, Hispanic, and Multi-racial were enrolled. After a 12 h fast, participants provided a saliva sample for the criterion baseline TBW measure and completed two bioelectrical impedance device (BIS, IB) measurements of TBW. Participants then ingested deuterium oxide (D2O). After an equilibration period, participants provided a second saliva sample for analysis of TBW.
Results:
For the total multi-ethnic sample, BIS estimates demonstrated good to fairly good agreement (Total error [TE] = 2.56 L, R2 = 0.951) and IB estimates were excellent to very good (TE = 1.95 L, R2 = 0.975). Validity results did not vary meaningfully between race and ethnicity.
Conclusions:
The results suggest the BIS and IB devices evaluated can produce valid TBW estimates compared to D2O in a multi-ethnic sample. TBW estimates from the IB may have better agreement with D2O compared to the BIS when using the default settings.
Keywords:
Agreement; Bioimpedance; Body composition; Deuterium oxide; Ethnicity; Racial groups.