点击显示 收起
1 From the National Institute of Child Health and Human Development, Bethesda, MD.
2 Address reprint requests to JA Yanovski, Unit on Growth and Obesity, Developmental Endocrinology Branch, National Institute of Child Health and Human Development, National Institutes of Health, 10 Center Drive, MSC 1862, Building 10, Room 10N262, Bethesda, MD 20892-1862. E-mail: jy15i{at}nih.gov.
See corresponding article on page 308.
One of the Holy Grails of obesity research has been the identification of groups of individuals whose unusually low resting energy expenditure (REE) predisposes them to weight gain. African American women have a higher risk of major weight gain than do white women (1) and have more difficulty losing weight (2). African American children also have a greater risk of overweight (3). As a result, the REEs of African Americans have been studied by many investigators. A recent review by Gannon et al (4) identified >15 studies that compared daily REE or total daily energy expenditure (EE) in African Americans and whites; about two-thirds of these studies showed lower EE [by between 363.7 and 1145.3 kJ (87 and 274 kcal)/d] in African Americans.
Most of the data that suggest that there are differences in REE between African Americans and whites were collected from adults; the picture has been somewhat less clear in cross-sectional studies of children. Morrison et al (5) reported lower REE in African American girls who fasted for 3 h before measurement of REE, and Kaplan et al (6) reported lower REE (adjusted for lean body mass) in African American children when lean body mass was determined by bioelectrical impedance analysis. In children who were studied under more controlled conditions, and using dual-energy X-ray absorptiometry or another precise method to determine body composition, we (7) and others (8–10) showed that African American children have a lower adjusted REE than do comparable white children. By contrast, one group of investigators reported on 4 different occasions (11–14) that REE was not different between African American and white children.
In this issue of the Journal, authors from this latter group of investigators present the first longitudinal analysis of REE to follow young children prospectively over a period long enough to observe the effects of pubertal development on EE. In contrast with the results of their previous cross-sectional studies, Sun et al (15) report that REE was significantly lower [by 172 kJ (41.2 kcal)/d, after adjustment for age, Tanner stage, fat mass, and lean body mass] in African American children than in white children.
How can the disparate cross-sectional and longitudinal data from the same group be reconciled with the data from other groups? The longitudinal study presented in this issue of the Journal suggests 2 possible answers. First, when relatively unselected populations of African American and white children are studied, the need to account for the myriad variables that may affect childhood EE requires a large sample. Inclusion of too many covariates may lead to a type 2 statistical error, even when a sizable number of children is studied. The present study reports a total of 566 repeated measurements of REE over a 5-y period from one of the largest groups ever studied: 92 white and 64 African American children. The second possible answer may be of equal importance in reconciling the data. In the present study, Sun et al (15) report a previously unknown observation: REE decreases as puberty progresses, after adjustment for ethnicity, sex, fat mass, and lean body mass. The magnitude of this effect [-315 kJ (75.4 kcal)/d for children at Tanner stage 3 and –935 kJ (223.7 kcal)/d for children at Tanner stage 4] exceeds the difference in REE they found between African American and white children. Thus, REE appears to join the ever-growing company of physiologic variables that change during the "Sturm und Drang" of puberty. Many humoral factors change temporarily during puberty, then resolve once adolescence is completed. Examples are increased growth hormone (16) and insulin (17) concentrations in midpubertal children. Given that there were only 59 observations in children at Tanner stage 3 and only 29 observations in children at Tanner stages 4 or 5 in the study by Sun et al (compared with 475 Tanner stage 1 and 144 Tanner stage 2 observations), these results remain somewhat uncertain, particularly given that previous cross-sectional studies did not show similar changes in REE with advancing puberty. Furthermore, it is not entirely clear from the description of the method that these investigators distinguished adequately between pubic hair stage and testicular or breast development stage (18, 19). However, should they be replicated, Sun et al's findings of lower REE in children at more advanced pubertal stages may prove to be another important confound affecting both the magnitude and the likelihood of detecting differences in REE between African American and white children. Other longitudinal studies of REE in adolescents should be carried out to verify whether there is a temporary decrease in REE that remits once adolescents complete their development.
Regardless of the effects of puberty, it seems that the preponderance of evidence now suggests that African American children have lower REE than do whites, even if the magnitude of the difference proves to be relatively small. The second aspect of the EE Holy Grail—that lower REE in African American children leads to a greater propensity for weight gain—remains unproven.
REFERENCES