The use of whole body calorimetry to compare measured versus predicted energy expenditure in postpartum women

Am J Clin Nutr. 2019 Mar 1;109(3):554-565. doi: 10.1093/ajcn/nqy312.


Background: Accurate assessment of energy expenditure may support weight-management recommendations. Measuring energy expenditure for each postpartum woman is unfeasible; therefore, accurate predictive equations are needed.

Objectives: This study compared measured with predicted resting energy expenditure (REE) and total energy expenditure (TEE) in postpartum women.

Methods: This was a longitudinal observational study. REE was measured at 3 mo postpartum (n = 52) and 9 mo postpartum (n = 49), whereas TEE was measured once at 9 mo postpartum (n = 43) by whole body calorimetry (WBC). Measured REE (REEWBC) was compared with 17 predictive equations; measured TEE plus breast milk energy output (ERWBC) was compared with the estimated energy requirements/Dietary Reference Intakes equation (EERDRI). Fat and fat-free mass were measured by dual-energy X-ray absorptiometry. Group-level agreement was assessed by the Pearson correlation, paired t test, and Bland-Altman (bias) analyses. Individual-level accuracy was assessed with the use of Bland-Altman limits of agreement, and by the percentage of women with predicted energy expenditure within 10% of measured values ("accuracy").

Results: The cohort was primarily Caucasian (90%). At a group level, the best equation predicting REEWBC was the DRI at 3 mo postpartum (-7 kcal, -0.1%; absolute and percentage bias, respectively), and the Harris-Benedict at 9 mo postpartum (-17 kcal, -0.5%). At an individual level, the Food and Agriculture Organization/World Health Organization/United Nations University (FAO/WHO/UNU) height and weight equation was the most accurate at 3 mo postpartum (100% accuracy) and 9 mo postpartum (98% accuracy), with the smallest limits of agreement. Equations including body composition variables were not more accurate. Compared with ERWBC, EERDRI bias was -36 kcal, with inaccurate predictions in 33% of women.

Conclusions: Many REE predictive equations were accurate for group assessment, with the FAO/WHO/UNU height and weight equation having the highest accuracy for individuals. EERDRI performed well at a group level, but inaccurately for 33% of women. A greater understanding of the physiology driving energy expenditure in the postpartum period is needed to better predict TEE and ultimately guide effective weight-management recommendations.

Keywords: body composition; breast milk energy output; estimated energy requirements; indirect calorimetry; lactation; obesity; postpartum women; predictive equations; resting energy expenditure; total energy expenditure.

Publication types

  • Comparative Study
  • Observational Study
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adult
  • Body Composition
  • Calorimetry, Indirect
  • Cohort Studies
  • Energy Metabolism*
  • Female
  • Humans
  • Longitudinal Studies
  • Postpartum Period / metabolism*

Grant support