Inter- and intraindividual correlations of background abundances of 2H, 18O and 17O in human urine and implications for DLW measurements

E. S. F. Berman*, E. L. Melanson, T. Swibas, S. P. Snaith, J. R. Speakman

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

19 Citations (Scopus)


BACKGROUND/OBJECTIVES: The method of choice for measuring total energy expenditure in free-living individuals is the doubly labeled water (DLW) method. This experiment examined the behavior of natural background isotope abundance fluctuations within and between individuals over time to assess possible methods of accounting for variations in the background isotope abundances to potentially improve the precision of the DLW measurement.

SUBJECTS/METHODS: In this work, we measured natural background variations in H-2, O-18 and O-17 in water from urine samples collected from 40 human subjects who resided in the same geographical area. Each subject provided a urine sample for 30 consecutive days. Isotopic abundances in the samples were measured using Off-Axis Integrated Cavity Output Spectroscopy.

RESULTS: Autocorrelation analyses demonstrated that the background isotopes in a given individual were not temporally correlated over the time scales of typical DLW studies. Using samples obtained from different individuals on the same calendar day, cross-correlation analyses demonstrated that the background variations of different individuals were not correlated in time. However, the measured ratios of the three isotopes H-2, O-18 and O-17 were highly correlated (R-2 = 0.89-0.96).

CONCLUSIONS: Although neither specific timing of DLW water studies nor intraindividual comparisons were found to be avenues for reducing the impact of background isotope abundance fluctuations on DLW studies, strong inter-isotope correlations within an individual confirm that use of a dosing ratio of 8%: 1% (0.6 p.p.m.: 1 p.p.m.) optimizes DLW precision. Theoretical implications for the possible use of O-17 measurements within a DLW study require further study.

Original languageEnglish
Pages (from-to)1091-1098
Number of pages8
JournalEuropean Journal of Clinical Nutrition
Issue number10
Early online date25 Mar 2015
Publication statusPublished - Oct 2015

Bibliographical note

We thank Ruixin Guo for statistical expertise, Susan Fortson for laboratory assistance and Thomas Owano for helpful discussions. We are particularly grateful to Professor Bill Wong for providing us with some enriched isotopic standards. This work was supported by NIH SBIR 1R43DK093362–01.


  • doubly-labeled water
  • cavity-output spectroscopy
  • examination-survey NHANES
  • energy-expenditure
  • physical-activity
  • CO2 production
  • laser spectroscopy
  • national-health
  • body-water
  • precision


Dive into the research topics of 'Inter- and intraindividual correlations of background abundances of 2H, 18O and 17O in human urine and implications for DLW measurements'. Together they form a unique fingerprint.

Cite this