Reproducibility and validity of bioimpedance spectroscopy for tracking changes in total body water: implications for repeated measurements
 
 
Journal Title (Medline/Pubmed accepted abbreviation): Br J Nutr
Year: 2010
Volume: 104
Number:
Page numbers: 1384-1394
doi: 10.1017/S0007114510002254

Summary of Background and Research Design

Background:Total body water measurements are necessary to assess body composition changes from diet and exercise. Dilution techniques such as 2H analysis produce the most accurate results, but are expensive, time consuming, and require fixed equipment. Bioimpedance spectroscopy (BIS) is a portable alternative technique with advances in technique and computing parameters that have allowed a fairly accurate assessment of total body water. However, BIS has limitations in its methodology and biologic factors that influence the results. Therefore, BIS may require larger changes in water volume to be meaningful compared with 2H analysis. Strict pretesting guidelines may reduce errors from biologic factors, but methodology errors still exist. One source of these errors is the reproducibility of the 4 single-site electrodes on the wrist and ankle; a 1-cm difference in electrode placement may affect resistance values by 2%. Fixed-distance electrodes may reduce the reproducibility errors.

Hypothesis/purpose of study: The purpose of this study was to compare the reproducibility of BIS total body water estimations between fixed-distance and single-site electrodes and to compare the validity and accuracy of fixed-distance and single-site electrodes with 2H analysis for tracking total body water changes after exercise.

Subjects: Twenty-nine healthy participants (15 men, 14 women) completed the reproducibility testing. Subjects were from 18 to 27 years old, with a mean body mass of 79 kg for men and 60 kg for women. Sixty-nine sedentary (< 30 min physical activity/wk), healthy participants (33 men, 36 women) completed the validity testing. Ages were from 18 to 45 years, with similar mean body mass for women in the exercise and no exercise (control) groups. There was a significant difference in mean body mass between men in the exercise vs control group (95.5 vs 83.0 kg; P < .05).

Experimental design: Open cohorts

Treatments and protocol: The reproducibility tests were conducted on 2 separate days, 24 hours apart. Participants fasted and avoided exercise for 12 hours before each test. Participants also wore the same clothes for each test. The BIS fixed-distance and single-site electrodes were the same size and shape and measurements were obtained according to the manufacturer’s recommendations. The fixed-distance electrodes were 5 cm apart (measured by a connecting strip between electrodes). The average of the 2 tests determined total body water for each participant. The validity tests (BIS and 2H analysis) were conducted during weeks 1 and 12 on the same day, with 12 hours between the 2 tests. The same pretesting conditions as the reproducibility tests were required. For weeks 1 to 12, participants choose between the exercise and control groups. The exercise regimen consisted of progressive endurance training (cycle ergometers 3 days/wk) and resistance training (9 single-joint and multi-joint exercises 2 days/wk, with 24 hr of recovery between sessions). The BIS testing was conducted the same as the prior testing, and the 2H analysis was conducted according to the standard procedure (2H ingestion with an empty bladder and a urine sample at 4 hr postingestion).

Summary of research findings:
  • Reproducibility testing demonstrated
    • Lower standard error of measurement in men with fixed-distance compared with single-site electrodes.
    • Identical measurements in women by either method.
    • Regression analyses showed no influence of body mass index, height, body mass, and total body water on the day-to-day errors of BIS.
    • Raw resistance values had lower standard errors of measurement, smaller mean differences, and lower minimal differences needed to be considered meaningful with fixed distance electrodes compared with single-site electrodes.
      • However, when comparing the final resistance values used to calculate total body water, there appeared to be no agreement between raw resistance values and the final reproducibility results.
  • Validity testing demonstrated
    • Fixed-distance electrodes increased total body water estimates compared with single-site electrodes (men, +0.872 L and women, +0.606 L).
    • Delta validation results showed no differences between electrode placements in total body water values (P = .08).
    • Significantly larger raw resistance values were observed from the calculated model at both zero and infinite frequencies from the fixed-distance electrodes compared with the single-site electrodes during both pre and post visits (P < .001).
    • Total body water was underestimated in women who exercised for both electrode placements compared with 2H analysis (P < .05).
    • No difference between BIS and 2H analysis was noted for the other groups.

Interpretation of findings/Key practice applications:

Regardless of electrode placement, BIS accurately tracked changes in total body water for the majority of participants, with a slightly better precision using the fixed-distance placement ( moderately decreased total body water reproducibility errors (day-to-day) compared with single-site electrodes). Because of the differing total body water reproducibility results for men and women, it may be necessary to report reproducibility by gender when using the fixed-distance electrodes. Additionally, reproducibility should be considered when interpreting group changes in total body water estimated by BIS, and results that include subjects who have not met the minimal differences needed to be considered meaningful should be interpreted with caution. Because both electrode placements resulted in valid delta results, either the single-site or the fixed-distance electrode position can provide accurate BIS total body water estimations. When taking multiple measurements to track changes in total body water, a fixed-distance electrode may provide accurate results with less change required to observe a meaningful difference. However, if a single measurement is required, the single-site electrode position is recommended until further research is conducted on the fixed-distance electrode position.
Google Tracking Google Plus Tracking Twitter Tracking