AJCN EB Program 2010
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Chomtho, S.
Right arrow Articles by Fewtrell, M. S
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Chomtho, S.
Right arrow Articles by Fewtrell, M. S
Agricola
Right arrow Articles by Chomtho, S.
Right arrow Articles by Fewtrell, M. S
American Journal of Clinical Nutrition, Vol. 88, No. 4, 1040-1048, October 2008
© 2008 American Society for Nutrition

ORIGINAL RESEARCH COMMUNICATION

Associations between birth weight and later body composition: evidence from the 4-component model1,2,3

Sirinuch Chomtho, Jonathan CK Wells, Jane E Williams, Alan Lucas and Mary S Fewtrell

1 From the Medical Research Council, Childhood Nutrition Research Centre, Institute of Child Health, University College London

2 Supported by the Medical Research Council (United Kingdom). SC was supported by the Anandamahidol Foundation and Chulalongkorn University Hospital.

3 Reprints not available. Address correspondence to S Chomtho, MRC Childhood Nutrition Research Centre, Institute of Child Health, 30 Guilford Street, London WC1N 1EH United Kingdom. E-mail: s.chomtho{at}ich.ucl.ac.uk.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 APPENDIX A
 REFERENCES
 
Background: Higher birth weight is associated with higher body mass index, traditionally interpreted as greater fatness or obesity, in later life. However, its relation with individual body-composition components and fat distribution remains unclear.

Objective: We investigated associations between birth weight and later fat mass (FM), fat-free mass (FFM), and fat distribution.

Design: Body composition was assessed by the criterion 4-component model in 391 healthy children [mean (±SD) age, 11.7 ± 4.2 y; 188 boys]. FM and FFM were adjusted for height (FMI = FM/height2; FFMI = FFM/height2) and were expressed as SD scores (SDS). Findings were compared between the 4-component and simpler methods.

Results: Birth weight was positively associated with height in both sexes and was significantly positively associated with FFMI in boys, equivalent to a 0.18 SDS (95% CI: 0.04, 0.32) increase in FFMI per 1 SDS increase in birth weight. These associations were independent of puberty, physical activity, social class, ethnicity, and parental body mass index. Birth weight was not significantly related to percentage fat, FMI, or trunk FMI in either sex. Equivalent analyses using simpler methods showed a trend for a positive relation between birth weight and FMI in boys that became nonsignificant after adjusting for confounders.

Conclusions: FFMI in later life in males is influenced by birth weight, a proxy for prenatal growth, but evidence for fetal programming of later FM or central adiposity is weak. Different body-composition techniques and data interpretation can influence results and should be considered when comparing studies.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 APPENDIX A
 REFERENCES
 
Research on the long-term effects of growth patterns in early life on later health, including obesity and cardiovascular disease, has increased rapidly during the past 15 y, with publications from both developed and developing countries. The term programming has been proposed for the concept that experience during early life is predictive of later health outcomes (1). Low birth weight has been linked to type 2 diabetes, cardiovascular disease, and the metabolic syndrome later in life (2–6). However, an extensive systematic review has shown that there is a positive association between birth weight and later body mass index (BMI) and hence, obesity, with each kilogram increase in birth weight typically increasing adult BMI by 0.5–0.7 kg/m2 (7). Although a higher BMI is traditionally interpreted as indicating greater fatness, it also reflects lean mass or FFM, and the limitations of BMI as an index of adiposity are increasingly appreciated. For instance, individual children of the same age and sex can have a 2-fold range of fat mass (FM) for a given BMI (8). Moreover, BMI also fails to reflect body fat distribution, which might be more important for development of the metabolic syndrome.

More recently, several studies have explored the relation between birth weight and later body composition. These studies differ in the age range, body-composition measurement techniques, and statistical approaches for analyzing body-composition data. However, several studies have shown consistent results in demonstrating positive associations between birth weight and later fat-free mass (FFM) in prepubertal children (9–11), adolescents (12, 13), and adults (14–16). In contrast, they have been much less consistent for FM and fat distribution, with studies variously reporting negative (4, 14–17), positive (10, 13) or nonsignificant associations with birth weight (9, 12). The inconsistency may partly reflect different body-composition data interpretation and measurement techniques. For example, most studies only used anthropometry or relied on a simple body-composition measurement such as bioelectrical impedance analysis. Only one recent study (17) used a reference method such as the four-component (4C) model to explore this issue despite the fact that body composition may play an important role in the programming of later disease risk, either itself being programmed by early growth or by being a mediator of the programming process. For example, Singhal et al (12) proposed the hypothesis that low birth weight might increase later disease risk by programming a smaller proportion of lean mass later in life, whereas other authors have suggested that low birth weight might program a tendency for central fat deposition (13, 16).

The primary aim of our study was to explore the relation between birth weight and later body composition by using the criterion 4C model in a population of UK children and adolescents. A secondary aim was to compare the results with those obtained with simpler models as well as highlighting important issues in the expression of body-composition data.


   SUBJECTS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
RESULTS
 DISCUSSION
 APPENDIX A
 REFERENCES
 
Healthy, term-born children and adolescents aged 4–20 y were recruited for a body-composition reference study with use of the 4C model at the Institute of Child Health and Great Ormond Street Hospital, London, United Kingdom. Recruitment was undertaken through advertisements in schools and sport clubs, the intranet, local newspapers, and word of mouth. The studied population was based in greater London and Cambridgeshire. All children were born full-term, were singletons, and did not have any known medical condition likely to affect growth or body composition. Ethical permission was obtained from the Research Ethics Committee of the Institute of Child Health and Great Ormond Street Hospital. Written informed consent was obtained from parents or from subjects older than 16 y, with written assent from children aged 11–15 y and verbal assent from the younger children.

Birth weight and gestational age data were collected from parental recall and were verified from the parent-held baby book where available (59.8%). There was no significant difference between parental recall and the baby book record for birth weight (mean difference –0.002 kg; 95% CI –0.014, 0.010). Birth weight SDS was calculated according to gestation and sex by using the British 1990 reference (18).

Measurement of body composition with the four-component model
The 4C model divides the body into fat, water, protein, and mineral as described previously (19, 20). It minimizes the assumptions made in the 2C model by directly evaluating several presumed constant relations (eg, hydration, density, and bone mineral content of FFM) that are fundamental to the 2C model. It has been accepted as the most robust technique for detecting interindividual variability in the composition of FFM and for showing consistent accuracy across a range of body fat (21). It might be assumed that combining several raw errors would result in poor precision for multicomponent values for body composition. However, this is not the case, as we discussed in detail previously (20). Thus, the 4C model, when using body volume measurements from the Bod Pod air-displacement plethysmograph (Life Measurement Inc, Concord, CA), has the best precision of all techniques and therefore is least likely to generate findings as an artifact of measurement error. This is because each individual error applies only to the component of weight that a given technique measures; hence, their sum remains small. We further discussed the effect of body size on this error and showed that the 4C model has precision of 0.22 kg in a child and 0.30 kg in an adult.

The various assumed densities of the 4 components were taken into account when calculating FM from the basic measurements.

Formula 1(1)
Where BV is body volume in L (from air-displacement plethysmography), TBW is total body water in L (from deuterium dilution), BMC is total-body bone mineral content in kg [from dual-energy X-ray absorptiometry (DXA)], and BW is body weight in kg.

FFM was then calculated as the difference between body weight and FM. To calculate the 4C model, measurements were obtained by using the following methods:

Air-displacement plethysmography
Whole-body air-displacement plethysmography was performed by using the Bod Pod body-composition system (Life Measurement Inc) with the subject wearing a tight-fitting swimming costume and a swimming cap. The machine provides raw body volume (L) for each subject from the difference between the volumes of air in this chamber, with and without the subject being present. Actual body volume was obtained after correction for thoracic gas volume and the surface area artifact by using the appropriate prediction equation for children from age, sex, weight, and height as described previously (22).

Deuterium dilution
Total body water was determined by 2H-labeled water dilution with a dose equivalent to 0.05 g of 2H2O (99.9%) per kg body weight. Saliva samples were obtained before and 4 h after dosing by use of absorbent salivettes at least 30 min after the last ingestion of food or drink, stored frozen at –30 °C, and then analyzed in duplicate using the equilibration method (23) and isotope-ratio mass spectrometry (Delta plus XP, Thermofisher Scientific, Bremen, Germany). For calculating total body water, it was assumed that the 2H2O dilution space overestimated total body water by a factor of 1.044 (24). Correction was made for dilution of the dose by water intake during the 4-h equilibration period.

Dual-energy X-ray absorptiometry
Bone mineral content, FM, and FFM were determined by using a GE Lunar Prodigy whole-body scanner (GE Medical Systems, Madison, WI) in conjunction with Encore 2002 software. The instrument automatically alters scan depth depending on the thickness of the subject, as estimated from age, height, and weight. A whole-body scan was performed while the subjects were wearing light indoor clothing and no removable metal objects. The typical scan duration was 5–10 min, depending on the subject's height. The radiation exposure per whole body scan was estimated to be 2.2 µSv, which is lower than the daily background radiation in the United Kingdom. All scans were performed and analyzed by one operator. The CVs (%) for a Lunar DPX-L instrument (regarded by the manufacturers to be similar to the Lunar Prodigy) have been reported as 1.10%, 2.0%, and 1.11% for total-body bone mineral content, FM, and lean mass, respectively (25). Regional measurements (arm, leg, and trunk) were marginally less precise than total body measurements, with CVs in the range of from 1% to 3%.

Because of its limitations in measuring soft tissue composition (26, 27), only total-body bone mineral content was used as a part of the 4-component model. However, for regional body composition, DXA is still a practical tool when compared with gold standards such as magnetic resonance imaging and computed tomography, which are more complicated, expensive, and, in the case of computed tomography, involve more radiation exposure. Therefore, DXA-derived trunk and limb composition were also used in some analyses.

Anthropometry
Body weight was measured with electronic scales to the nearest 0.01 kg, and height was measured to the nearest 0.1 cm with a wall-mounted stadiometer (Holtain, Dyfed, United Kingdom). BMI was calculated as weight (kg) divided by the square of height (m2). Waist circumference was measured at the natural waist site (28) with a nonstretchable, fiberglass insertion tape to the nearest 0.1 cm. All measurements were undertaken by 1 of the 4 trained investigators.

Confounding variables
Factors that might confound the relation between birth weight and body composition were assessed by using a structured questionnaire. Pubertal status was self-assessed with the use of pictures of the Tanner stages for pubic hair and breast (female) or genital (male) development and was coded as prepubertal (Tanner stage 1), early pubertal (Tanner stages 2 and 3), or late pubertal (Tanner stages 4 and 5). A simple parental assessment of the subjects’ physical activity level relative to their peers was made by use of a 5-point scale ranging from much less active than peers, less active than peers, same as peers, more active than peers, and much more active than peers. Social class was assessed by using the Standard Occupational Classification (29) and was classified as class 1 (high), 2, 3 or ≥4. Ethnicity was coded as white or nonwhite, because in this dataset there were too few subjects for more detailed analyses of the nonwhite group. Reported parental height and weight were used to calculate parental BMI.

Statistical methods
SD scores (SDS) were derived for weight, height, and BMI by using the 1990 British reference (30, 31) and were calculated by use of the lmsGrowth program (copyright 2002–2005, Medical Research Council, London, United Kingdom). Waist circumference was converted to SDS by using a nationally representative sample of UK children in 1988 (28) and was calculated by use of the lmsGrowth program. One-sample t tests were used to compare the SDS of the study population with reference data.

We adjusted all body-composition variables by height squared to adjust for body size in a way comparable to that used for BMI (32). Total and regional FM index (FMI, FM/height2) and FFM index (FFMI, FFM/height2) were then calculated. The body-composition variables were converted to age- and sex-specific SDS by using our own reference database of 449 subjects aged 4–23 y (204 boys) from an ongoing body-composition reference study (33). The SDS were calculated by using the LMS method (34) and the lmsChartMaker program (Copyright 1997–2005, Medical Research Council). This method also allows direct comparison of the effect size of early growth on later FM or FFM because they are both in SDS, thus avoiding the problem of unequal variance between different body-composition variables.

The association between birth weight and later body composition was explored by using multiple regression models. Body-composition outcomes were regressed on birth weight SDS with adjustment for other confounders. Alternatively, birth weight SDS were divided into quartiles and the adjusted mean of body-composition outcomes for each birth weight SDS quartile were analyzed by general linear models. General linear models were also used to test the interactions between birth weight SDS and sex or pubertal status in predicting later body composition. All analyses were performed by using SPSS version 13 (SPSS Inc, Chicago).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
DISCUSSION
 APPENDIX A
 REFERENCES
 
Characteristics of the study subjects
Descriptive statistics of the subjects (age range: 4.2–20.2 y) are summarized in Table 1Go. Mean (±SD) gestational age was 40.1 ± 1.4 wk (range: 36–43 wk), and mean birth weight was 3.47 ± 0.51 kg (range: 2.01–5.08 kg), with no statistically significant difference in gestational age or birth weight between boys and girls. Gestation and sex-specific birth weight SDS (range: –3.31 to 3.47) were comparable with the British 1990 reference for both sexes.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Characteristics of the study subjects1

 
Boys and girls in this dataset were taller, heavier, and had higher BMIs than the British 1990 reference. Compared with reference data (in the form of SDS), both sexes also had higher waist circumferences. Girls also had higher waist circumference SDS than did boys. In the subgroup analysis (data not shown), this difference was significant only in pubertal subjects. In general, boys had lower FMI and higher FFMI (both total and regional) than did girls. Because body composition differed according to sex with a significant interaction between birth weight and sex in predicting BMI, waist circumference, FFMI, trunk FFMI, and limb FFMI (P for interaction 0.03, 0.04, 0.01, 0.01, and 0.03, respectively), all subsequent analyses were performed separately for boys and girls. There were no sex differences in confounders (parental BMI, ethnicity, pubertal status, physical activity, and social class) between boys and girls.

Relation of potential confounders with body-composition outcomes
We examined associations between each potential confounding factor and both birth weight and later body-composition variables (Table A1Go in Appendix A). Maternal BMI was positively associated with birth weight SDS. A lower social class was related to lower birth weight SDS in girls only. These findings are important because maternal BMI and social class were also related to later body-composition outcomes.


View this table:
[in this window]
[in a new window]

  		TABLE A1. Correlations (r) between potential confounders and birth weight and body-composition outcomes (SDS form)1

 
Puberty showed a significant negative correlation with FMI SDS in boys and a significant positive association with FFMI SDS in girls. Physical activity was significantly negatively associated with FMI in both sexes and positively associated with FFMI in girls. Social class also showed a significant association with FMI in boys (tendency for greater fatness with lower social class) and with FFMI in girls (tendency for higher FFMI with lower social class). Maternal BMI had a stronger positive correlation with offspring FMI and FFMI than did paternal BMI, for which the relation was significant only in boys. We adjusted for these confounders in subsequent multiple regression analyses.

Birth weight and later body size and composition
The results of multiple regression analyses are shown in Table 2Go (unadjusted models) and Table 3Go (adjusted models). Birth weight was positively associated with later height in both sexes; at any given age, a 1-SDS increase in birth weight corresponded with 0.20 and 0.25 SDS increases in later height (after adjustment for confounders) in boys and girls, respectively (95% CI: 0.06–0.34 for boys, 0.09–0.41 for girls). However, this association was attenuated substantially and was no longer statistically significant (P = 0.051 and 0.07) if parental height was used instead of parental BMI (data not shown). Birth weight SDS positively predicted later BMI SDS only in boys; a 1-SDS increase in birth weight predicted a 0.20-SDS increase (95% CI: 0.02–0.38) in later BMI after adjustment for potential confounders.


View this table:
[in this window]
[in a new window]

  		TABLE 2. Regression of current body size and body composition (BC) on birth weight SD score (SDS): unadjusted model1

 

View this table:
[in this window]
[in a new window]

  		TABLE 3. Regression of current body size and body composition (BC) on birth weight SD score (SDS): adjusted for puberty, physical activity, social class, ethnicity, and parental BMI1

 
The relation between birth weight and later FFMI differed according to sex (P for interaction = 0.01), with a positive association in boys, equivalent to around a 0.18-SDS increase (95% CI: 0.04, 0.32) in FFMI per 1-SDS increase in birth weight. The relation was unaffected by adjustment for confounders. In girls, there was no association. The association between birth weight and regional FFMI from DXA (trunk and limb FFMI) was consistent with the findings for total FFMI from the 4C model in both sexes. The proportion of the variance in later FFMI explained by birth weight was small ({approx}3%) but comparable with that explained by maternal BMI (2%), paternal BMI (4.1%), or current activity level (3%) in a multivariate model.

Birth weight was not significantly related to FMI or percentage fat (expressed as an SD score) in either sex, and there was no evidence that the associations for FMI differed between boys and girls (P for interaction = 0.6). There was also no significant association between birth weight and absolute percentage fat in either sex, either unadjusted or after adjustment for confounders (girls, P = 0.8; boys, P = 0.3).

Birth weight and fat distribution
Regression analyses of different proxies for fat distribution on birth weight SDS are shown in Table 2Go. In boys, there was a positive association between birth weight and waist circumference; equivalent to around a 0.20-SDS increase (95% CI: 0.04, 0.36) in waist circumference per 1-SDS increase in birth weight. This association remained in the same direction but was attenuated after FMI SDS was added to the model. There was also a positive association between birth weight and trunk FMI that became nonsignificant after adjustment for nontrunk (limb) FMI. As shown in Table 3Go (adjusted model), these associations were substantially attenuated and became nonsignificant after adjustment for relevant confounders. In girls, birth weight was not associated with either waist circumference or trunk FMI in any of the models.

Effects of pubertal status
There was no statistically significant interaction between birth weight and pubertal status in predicting any of the body-composition outcomes. However, subgroup analyses in boys showed a significant positive association between birth weight SDS and FFMI in prepubertal boys (n = 87), whereas the relation was not significant in early or late pubertal boys (n = 100) (data not shown). In prepubertal boys (n = 87), birth weight was also negatively associated with percentage fat, although only after adjustment for BMI and other confounding factors (β = –0.9, P = 0.047). There was no such association in prepubertal girls.

Effects of using different body-composition measurement techniques
Because most published studies have used 2C techniques, typically skinfold-thickness equations and DXA, to determine body composition in later life, we repeated the same analyses within our dataset by using FM and FFM derived from DXA and from the skinfold-thickness equation of Slaughter et al (35). There was a significant positive association between birth weight and FFMI in boys that remained after adjustment for confounders and no association in girls, which is consistent with results from the 4C model. However, using data from these 2C methods, there was also a trend for a positive relation between birth weight and FM in boys that became nonsignificant after adjustment for confounders (see Table A2 in Appendix A).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
APPENDIX A
 REFERENCES
 
We showed that higher birth weight was associated with greater height in both sexes and with higher BMI in boys. Higher birth weight was also associated with higher FFMI in boys but not girls; this association was seen from childhood to young adulthood but was more prominent in prepubertal boys. Associations were independent of current body size, age, pubertal stage, physical activity, social class, ethnicity, and parental BMI. The proportion of variance in later FFMI explained by birth weight was small, but comparable with that explained by parental BMI or current activity level. Although the sample size in our study was not comparable with that in larger epidemiologic studies, the main strength of our study was the use of the 4C model, which provided the most accurate body-composition measurements possible and showed some important points regarding the interpretation of such data.

The finding that birth weight is positively associated with measures of later FFM is supported by several studies [see a recent review (36)]. In contrast, the relation between birth weight and FM in published studies is less consistent, with several reporting a negative association between birth weight and proxies of fatness only after adjustment for current body size (weight or BMI) (13, 16, 17, 36). The only previous study to examine relations between birth weight and later body composition using the 4C model (17) reported a negative association between birth weight and percentage fat in prepubertal children, both with and without adjustment for confounders. For comparison, we also presented data on percentage fat and found no significant association with birth weight in the whole cohort, although our subgroup analysis in prepubertal boys showed a negative association, albeit only after adjustment for confounders, including BMI. However, as discussed elsewhere (32), we consider height to be a more appropriate parameter for size adjustment than weight (or measures derived from weight such as percentage body fat); the latter precludes individual consideration of fat and lean mass because one is the inverse of the other. A negative association between birth weight and later percentage fat or FM adjusted for weight or BMI could be interpreted as indicating that a lower birth weight is associated with either higher FM or lower FFM in later life, because a lower FFM with the same FM would result in higher percentage fat. In our dataset, when we used height to adjust for body size (without adjustment for later weight or BMI), there was no significant relation between birth weight and FM.

Interestingly, the previous study (17) that used the 4C model also reported that, with adjustment for age, sex, weight, and height, birth weight was significantly positively related to later FFM, but there was no significant association with later FM. Although not emphasized in the discussion, this is consistent with our own findings.

Most studies looking at the relation between birth weight and later central fat distribution used subscapular skinfold thickness, the subscapular-to-triceps ratio, waist circumference, or waist-to-hip ratio and reported a negative association after adjustment for current weight or BMI (36). This finding has generally been interpreted as indicating the programming of later central adiposity by low birth weight. However, alternative explanations are that postnatal growth or the prenatal constraint of growth (indicated by low birth weight) interacting with rapid postnatal growth, predict later central fatness. Moreover, we argued previously that the skinfold thickness ratio or waist-to-hip ratio may not be appropriate parameters for assessing relative fat distribution because of poor statistical validity (36, 37).

We found a positive relation between birth weight SDS and later waist circumference (but not trunk FM) in boys both before and after adjustment for total FMI, which seems counterintuitive. Waist circumference is a good indicator of metabolic risk in adults and children (38) but it is not a direct measure of visceral adipose tissue, the best predictor of the metabolic syndrome. Our interpretation of this finding is therefore that, among adolescent boys with the same height and total FM, those with a higher birth weight have a greater waist circumference partly because they have more trunk FFM, which may track from birth. Results from the limited number of studies that have investigated this issue using direct measures of central fatness (from DXA or CT) are inconsistent (4, 10, 39). This issue is difficult to resolve because of the lack of a valid method for measuring central fat distribution in large populations.

Effects of using different body-composition measurement techniques
In larger studies, anthropometry has been used to derive body fat from skinfold-thickness measurements. More recently, DXA has been used to assess body composition in children. Both methods involve several assumptions. We showed that if DXA or skinfold-thickness equation-derived FM was used in the analysis, there was a weak positive association between birth weight and FM that was not apparent when using data from the 4C model. However, our main finding of an association between birth weight and FFMI in boys did not change with the use of different body-composition techniques. The 4C model requires relatively complex measurements and specialized equipment and is therefore more difficult, expensive, and time-consuming to perform. In practice, one needs to consider the relative benefits of a small study with very accurate measurements and a larger study with simpler measurements. However, it is important to appreciate the limitations and assumptions of each body-composition technique used when interpreting the results. The validation of field methods against a more accurate method and the use of assumptions suitable for the population of interest are essential (40).

Sex differences in the relation between birth weight and body composition
An interaction between sex and birth weight in predicting later body composition is not well established. Few studies in prepubertal children considered males and females separately. In the ALSPAC cohort (10), the association between birth weight and FM or trunk FM did not differ by sex. However, there was an interaction between sex and birth size (ponderal index) in predicting FFM (P = 0.04), with a larger effect on lean mass in boys, which is consistent with our findings. A few studies have been conducted in adolescents, when sex differences in body composition are expected to be more pronounced. Singhal et al (12) found a positive association between birth weight and FFMI in both sexes. However, Labayen et al (13) reported a significant interaction between sex and birth SDS in predicting later FFM in 234 Spanish adolescents, with an association only in girls.

Mechanism for the programming of later FFM
There are several plausible explanations for the observed association between birth weight and later FFMI. In humans, the number of muscle fibers is set before birth (41) with little hyperplasia during postnatal life. Fetal growth constraint could result in lower insulin-like growth factor-I with compromised muscle growth and skeletal length in utero (42). There is indeed some evidence that small for gestational age infants show a larger deficit in lean mass (by DXA) than in FM (43), which might then persist into later life. It could also be argued that fetal muscle and skeletal growth is determined (at least in part) by an individual's genetic potential (44–46) and that this is responsible for the relation with later FFMI. This hypothesis cannot, however, explain the association between the intrapair difference in birth weight of monozygotic twins and later FFM described by Loos et al (14, 15). We propose that the programming of FFM involves insults or stimuli in fetal life, and to a lesser extent, early postnatal life. Catalano et al (46) reported that male neonates had higher FFM than females, which corresponds with the well-recognized sex differences in body composition during adolescence and adult life. This could also explain why the relation between birth weight and FFM was more obvious in boys.

Study limitations
Birth weights in our cohort were representative of the 1990 UK reference population, whereas, consistent with the secular trend in the United Kingdom, our subjects were taller and heavier than average at follow-up. At the time, 81 subjects (20.7%) were overweight and 17 (4.3%) were obese according to the International Obesity Task Force cutoff for BMI (47). Our findings should therefore be generalizable to similar contemporary populations. Our body-composition outcomes were measured over a wide range of age and stages of pubertal maturation, but we expressed body-composition variables as age- and sex-specific SDS, and our results did not change after adjustment for pubertal maturation in the models. The use of birth weight as a proxy for prenatal growth has some limitations for the study of body-composition programming. There is some evidence, for example, that body composition at birth may vary even when birth weight is the same and within the normal range (46). Yajnik et al (48) showed that Indian babies have a more adipose body composition despite being "small and thin" relative to UK infants.

In conclusion, our data suggest that FFMI in later life in males is influenced by birth weight, which is a proxy for prenatal growth; whereas the evidence for programming of later FM or central fat distribution by birth weight is not strong. A study with measurements of body composition at birth in relation to later outcomes might help to clarify the mechanisms underlying our observations. Body-composition measurement techniques and the methods used for data interpretation can influence results and should be considered when comparing studies.


   APPENDIX A
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 APPENDIX A
REFERENCES
 


Figure 1
View larger version (19K):
[in this window]
[in a new window]

  		FIGURE A1.. Adjusted mean SDS for later height, BMI, FMI, FFMI, trunk FMI, and trunk FFMI stratified according to birth weight SD score (SDS) quartiles. Error bars indicate 95% CI. The associations between birth weight SDS and body-composition outcomes were tested by general linear model and adjusted for puberty, physical activity, social class, ethnicity, and parental BMI. FMI, fat mass index; FFMI, fat-free mass index; WC, waist circumference.

 

View this table:
[in this window]
[in a new window]

  		TABLE A2. Regression of current body size and body composition (BC) derived from the 2-component models on birth weight SD score (SDS)1: unadjusted2 and adjusted3 model for puberty, physical activity, social class, ethnicity, and parental BMI

 


   ACKNOWLEDGMENTS
 
We thank the children and their parents for participating in the study and Catherine M Wilson for conducting and analyzing all DXA scans.

The contributions of the authors were as follows—SC and MSF designed the study and analyzed the data, JEW and SC measured the subjects and modeled the body-composition data, SC prepared the first draft of the manuscript, JCKW provided critical input on the body-composition protocol and data analyses, and AL provided advice on the discussion. All authors participated in the interpretation of the results and contributed to revision of the manuscript. None of the authors had any conflicts of interest.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 APPENDIX A
 REFERENCES
 

  1. Lucas A. Programming by early nutrition in man. In: Bock GR, Whelan J, eds. The childhood environment and adult disease. Chichester, United Kingdom: Wiley, 1991:38–55.
  2. Barker DJ, Hales CN, Fall CH, Osmond C, Phipps K, Clark PM. Type 2 (non-insulin-dependent) diabetes mellitus, hypertension and hyperlipidaemia (syndrome X): relation to reduced fetal growth. Diabetologia 1993;36:62–7.[Medline]
  3. Bavdekar A, Yajnik CS, Fall CH, et al. Insulin resistance syndrome in 8-year-old Indian children: small at birth, big at 8 years, or both? Diabetes 1999;48:2422–9.[Abstract]
  4. Garnett SP, Cowell CT, Baur LA, et al. Abdominal fat and birth size in healthy prepubertal children. Int J Obes Relat Metab Disord 2001;25:1667–73.[Medline]
  5. Barker DJ, Eriksson JG, Forsen T, Osmond C. Fetal origins of adult disease: strength of effects and biological basis. Int J Epidemiol 2002;31:1235–9.[Abstract/Free Full Text]
  6. Ramadhani MK, Grobbee DE, Bots ML, et al. Lower birth weight predicts metabolic syndrome in young adults: the Atherosclerosis Risk in Young Adults (ARYA)-study. Atherosclerosis 2006;184:21–7.[Medline]
  7. Parsons TJ, Power C, Logan S, Summerbell CD. Childhood predictors of adult obesity: a systematic review. Int J Obes Relat Metab Disord 1999;23(suppl 8):S1–107.
  8. Wells JC. A Hattori chart analysis of body mass index in infants and children. Int J Obes Relat Metab Disord 2000;24:325–9.[Medline]
  9. Wells JC, Hallal PC, Wright A, Singhal A, Victora CG. Fetal, infant and childhood growth: relationships with body composition in Brazilian boys aged 9 years. Int J Obes (Lond) 2005;29:1192–8.[Medline]
  10. Rogers IS, Ness AR, Steer CD, et al. Associations of size at birth and dual-energy X-ray absorptiometry measures of lean and fat mass at 9 to 10 y of age. Am J Clin Nutr 2006;84:739–47.[Abstract/Free Full Text]
  11. Murphy MJ, Metcalf BS, Jeffery AN, Voss LD, Wilkin TJ. Does lean rather than fat mass provide the link between birth weight, BMI, and metabolic risk? EarlyBird 23. Pediatr Diabetes 2006;7:211–4.[Medline]
  12. Singhal A, Wells J, Cole TJ, Fewtrell M, Lucas A. Programming of lean body mass: a link between birth weight, obesity, and cardiovascular disease? Am J Clin Nutr 2003;77:726–30.
  13. Labayen I, Moreno LA, Blay MG, et al. Early programming of body composition and fat distribution in adolescents. J Nutr 2006;136:147–52.[Abstract/Free Full Text]
  14. Loos RJ, Beunen G, Fagard R, Derom C, Vlietinck R. Birth weight and body composition in young adult men—a prospective twin study. Int J Obes Relat Metab Disord 2001;25:1537–45.[Medline]
  15. Loos RJ, Beunen G, Fagard R, Derom C, Vlietinck R. Birth weight and body composition in young women: a prospective twin study. Am J Clin Nutr 2002;75:676–82.[Abstract/Free Full Text]
  16. Kensara OA, Wootton SA, Phillips DI, Patel M, Jackson AA, Elia M. Fetal programming of body composition: relation between birth weight and body composition measured with dual-energy X-ray absorptiometry and anthropometric methods in older Englishmen. Am J Clin Nutr 2005;82:980–7.[Abstract/Free Full Text]
  17. Elia M, Betts P, Jackson DM, Mulligan J. Fetal programming of body dimensions and percentage body fat measured in prepubertal children with a 4-component model of body composition, dual-energy X-ray absorptiometry, deuterium dilution, densitometry, and skinfold thicknesses. Am J Clin Nutr 2007;86:618–24.[Abstract/Free Full Text]
  18. Cole TJ, Freeman JV, Preece MA. British 1990 growth reference centiles for weight, height, body mass index and head circumference fitted by maximum penalized likelihood. Stat Med 1998;17:407–29.[Medline]
  19. Fuller NJ, Jebb SA, Laskey MA, Coward WA, Elia M. Four-component model for the assessment of body composition in humans: comparison with alternative methods, and evaluation of the density and hydration of fat-free mass. Clin Sci (Lond) 1992;82:687–93.[Medline]
  20. Wells JC, Fuller NJ, Dewit O, Fewtrell MS, Elia M, Cole TJ. Four-component model of body composition in children: density and hydration of fat-free mass and comparison with simpler models. Am J Clin Nutr 1999;69:904–12.[Abstract/Free Full Text]
  21. Bray GA, DeLany JP, Volaufova J, Harsha DW, Champagne C. Prediction of body fat in 12-y-old African American and white children: evaluation of methods. Am J Clin Nutr 2002;76:980–90.[Abstract/Free Full Text]
  22. Wells JC, Fuller NJ, Wright A, Fewtrell MS, Cole TJ. Evaluation of air-displacement plethysmography in children aged 5-7 years using a three-component model of body composition. Br J Nutr 2003;90:699–707.[Medline]
  23. Scrimgeour CM, Rollo MM, Mudambo SM, Handley LL, Prosser SJ. A simplified method for deuterium/hydrogen isotope ratio measurements on water samples of biological origin. Biol Mass Spectrom 1993;22:383–7.[Medline]
  24. Racette SB, Schoeller DA, Luke AH, Shay K, Hnilicka J, Kushner RF. Relative dilution spaces of 2H- and 18O-labeled water in humans. Am J Physiol 1994;267:E585–90.[Medline]
  25. Kiebzak GM, Leamy LJ, Pierson LM, Nord RH, Zhang ZY. Measurement precision of body composition variables using the lunar DPX-L densitometer. J Clin Densitom 2000;3:35–41.[Medline]
  26. Laskey MA. Dual-energy X-ray absorptiometry and body composition. Nutrition 1996;12:45–51.[Medline]
  27. Williams JE, Wells JC, Wilson CM, Haroun D, Lucas A, Fewtrell MS. Evaluation of Lunar Prodigy dual-energy X-ray absorptiometry for assessing body composition in healthy persons and patients by comparison with the criterion 4-component model. Am J Clin Nutr 2006;83:1047–54.[Abstract/Free Full Text]
  28. McCarthy HD, Jarrett KV, Crawley HF. The development of waist circumference percentiles in British children aged 5.0-16.9 y. Eur J Clin Nutr 2001;55:902–7.[Medline]
  29. Standard occupational classification: a publication of the Government Statistical Service. London: HMSO, 1991.
  30. Freeman JV, Cole TJ, Chinn S, Jones PR, White EM, Preece MA. Cross sectional stature and weight reference curves for the UK, 1990. Arch Dis Child 1995;73:17–24.[Abstract/Free Full Text]
  31. Cole TJ, Freeman JV, Preece MA. Body mass index reference curves for the UK, 1990. Arch Dis Child 1995;73:25–9.[Abstract/Free Full Text]
  32. Wells JC, Cole TJ. Adjustment of fat-free mass and fat mass for height in children aged 8 y. Int J Obes Relat Metab Disord 2002;26:947–52.[Medline]
  33. Chomtho S. The influence of early nutrition and growth on body composition in childhood and early adult life. PhD thesis. 1-302. University College London, United Kingdom, 2006.
  34. Cole TJ, Green PJ. Smoothing reference centile curves: the LMS method and penalized likelihood. Stat Med 1992;11:1305–19.[Medline]
  35. Slaughter MH, Lohman TG, Boileau RA, et al. Skinfold equations for estimation of body fatness in children and youth. Hum Biol 1988;60:709–23.[Medline]
  36. Wells JC, Chomtho S, Fewtrell MS. Programming of body composition by early growth and nutrition. Proc Nutr Soc 2007;66:423–34.[Medline]
  37. Wells JC, Victora CG. Indices of whole-body and central adiposity for evaluating the metabolic load of obesity. Int J Obes (Lond) 2005;29:483–9.[Medline]
  38. Maffeis C, Pietrobelli A, Grezzani A, Provera S, Tato L. Waist circumference and cardiovascular risk factors in prepubertal children. Obes Res 2001;9:179–87.[Medline]
  39. Choi CS, Kim C, Lee WJ, et al. Association between birth weight and insulin sensitivity in healthy young men in Korea: role of visceral adiposity. Diabetes Res Clin Pract 2000;49:53–9.[Medline]
  40. Wells JC, Fewtrell MS. Measuring body composition. Arch Dis Child 2006;91:612–7.[Abstract/Free Full Text]
  41. Malina RM. Post-natal growth and maturation. In: Ulijaszek SJ, Johnston FE, Preece MA, eds. The Cambridge encyclopedia of human growth and development. Cambridge, United Kingdom: Cambridge University Press, 1998:176–235.
  42. Gluckman PD, Harding JE. Fetal growth retardation: underlying endocrine mechanisms and postnatal consequences. Acta Paediatr Suppl 1997;422:69–72.[Medline]
  43. Lapillonne A, Braillon P, Claris O, Chatelain PG, Delmas PD, Salle BL. Body composition in appropriate and in small for gestational age infants. Acta Paediatr 1997;86:196–200.[Medline]
  44. Godfrey K, Walker-Bone K, Robinson S, et al. Neonatal bone mass: influence of parental birthweight, maternal smoking, body composition, and activity during pregnancy. J Bone Miner Res 2001;16:1694–703.[Medline]
  45. Knight B, Shields BM, Turner M, Powell RJ, Yajnik CS, Hattersley AT. Evidence of genetic regulation of fetal longitudinal growth. Early Hum Dev 2005;81:823–31.[Medline]
  46. Catalano PM, Drago NM, Amini SB. Factors affecting fetal growth and body composition. Am J Obstet Gynecol 1995;172:1459–63.[Medline]
  47. Cole TJ, Bellizzi MC, Flegal KM, Dietz WH. Establishing a standard definition for child overweight and obesity worldwide: international survey. BMJ 2000;320:1240–3.[Abstract/Free Full Text]
  48. Yajnik CS, Fall CH, Coyaji KJ, et al. Neonatal anthropometry: the thin-fat Indian baby. The Pune Maternal Nutrition Study. Int J Obes Relat Metab Disord 2003;27:173–80.[Medline]
Received for publication November 14, 2007. Accepted for publication June 20, 2008.




This article has been cited by other articles:


Home page
 Am. J. Clin. Nutr.Home page
P. M. Skidmore, A. Cassidy, R. Swaminathan, J B. Richards, M. Mangino, T. D Spector, and A. J MacGregor
An obesogenic postnatal environment is more important than the fetal environment for the development of adult adiposity: a study of female twins
Am. J. Clinical Nutrition, August 1, 2009; 90(2): 401 - 406.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Chomtho, S.
Right arrow Articles by Fewtrell, M. S
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Chomtho, S.
Right arrow Articles by Fewtrell, M. S
Agricola
Right arrow Articles by Chomtho, S.
Right arrow Articles by Fewtrell, M. S

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS