Reproducibility of measurement of resting energy expenditure in prepubertal girls

We determined the reproducibility of measuring resting energy expenditure (REE) and the effect on REE of speeding the night before testing at home compared with in a clinical research center. We studied 19 prepubertal girls aged 6.0-10.1 y with a mean weight-for-height of 108% of ideal. REE was measured for 30 min with a metabolic monitor after 12 h of fasting on three consecutive mornings during two different hospitalizations 6 wk apart. The initial REE measurement of each hospitalization was obtained on admission and the second and third measurements were performed during the 3-d hospitalization. Energy intake was ad libitum and was recorded while the children were in the research center. Body composition was determined twice with dual-energy X-ray absorptiometry. No significant effect on REE of day within visit was found across visits. There was no significant difference between the initial REE measurement and the second and third measurements. The mean REEs for the two hospitalizations were highly correlated. The mean CV in intraindividual REE (ie, all six measurements) was 5.8% and was unchanged when adjusted for fat-free body mass (FFM) or body weight. The mean CV in interindividual REE decreased when adjusted for FFM or body weight. Because REE was highly reproducible, a single measurement can suffice for energy expenditure studies in girls aged 6-10 y. Admission to a clinical research setting is not necessary for a reliable determination of REE.