Excess weight gain during both pre- and postnatal life increases risk for obesity in later life. Although a number of gestational and early life contributors to this effect have been identified, there is a dearth of research to examine whether gestational factors and weight gain velocity in infancy exert independent effects on subsequent body composition and fat distribution.
To test the hypothesis that birth weight, as a proxy of prenatal weight gain, and rate of weight gain before 6 months would be associated with total and truncal adiposity at 12 months of age.
Design and Methods:
Healthy, term infants (N = 47) were enrolled in the study and rate of weight gain (g/day) was assessed at 0-3 months, 3-6 months, and 6-12 months.
Total and regional body composition were measured by dual-energy X-ray absorptiometry (DXA) at 12 months. Stepwise linear regression modeling indicated that lean mass at 12 months, after adjusting for child length, was predicted by rate of weight gain during each discrete period of infancy (P < 0.05), and by maternal pre-pregnancy BMI (P < 0.05). Total fat mass at 12 months was predicted by rate of weight gain during each discrete period (P < 0.01), and by older maternal age at delivery (P < 0.05). Trunk fat mass at 12 months, after adjusting for leg fat mass, was predicted by rate of weight gain from 0-3 months and 3-6 months (P < 0.05).
Results suggest that growth during early infancy may be a critical predictor of subsequent body composition and truncal fat distribution.
Excess weight gain during both pre- and postnatal life increases risk for obesity in later life. High birth weight, as a surrogate marker for intrauterine growth, is a positive predictor of obesity during childhood and adulthood (1,2). This effect of birth weight may be mediated by maternal pre-pregnancy obesity (3-,4,5), and/or excess gestational weight gain (GWG) (6,7). Beyond birth, studies have consistently shown that rapid weight gain in the first 6 months is also associated with overweight and obesity during childhood (5,8,9), adolescence (10,11), and adulthood (10,12,13). Furthermore, excess weight gain during the first, but not the second, 6 months of life, has been associated with relatively greater trunk fat among 4-20-year-old individuals (14). This is particularly important given the association between abdominal obesity and adverse metabolic health (for review see ref. (15,16)). There is a dearth of research that concurrently investigates the roles of both maternal gestational factors and weight gain during infancy on body composition and fat distribution during infancy.
The purpose of this study was to examine the independent effects of maternal pregestational BMI, GWG, infant birth weight, and the rate of infant weight gain during discrete periods of infancy on total fat and lean mass, and on truncal adiposity at 12 months of age. We hypothesized that both prenatal weight gain (as reflected in birth weight) and postnatal weight gain during the first 6 months of life would independently predict total and truncal adiposity at 12 months of age.
Methods and Procedures
Healthy full-term infants and their mothers were recruited from ongoing growth studies conducted at the University of Oklahoma Health Sciences Center. Inclusion criteria included maternal age of 18-45 years at delivery, singleton pregnancy, delivery at ≥37 weeks, hospital discharge of the infant not more than 3 days after delivery, infant weight and length measurements at 3, 6, and 12 months and child dual-energy X-ray absorptiometry (DXA) scan at 12 months (N = 51). Exclusion criteria were maternal pregestational or gestational diabetes, admission of tobacco or illicit drug use during pregnancy, consumption of >1 alcoholic drink/week during pregnancy, and infant with presumed or known congenital defects.
This study was conducted as a secondary analysis of data collected during ongoing early childhood growth studies at the University of Oklahoma. For all primary study protocols, mothers and infants came to the clinic when infants were 3 months ± 7 days, 6 months ± 7 days, and 12 months ± 14 days of age. Birth weight, length, and gestational age at delivery were obtained from hospital records. Birth weight for length z-score (BWLz) was calculated using gender-specific reference data from the Centers for Disease Control and Prevention Growth Charts (http://www.cdc.gov/growthcharts). Maternal race/ethnicity, parity, age at delivery, pre-pregnancy weight, and GWG, were based upon self-report. Pre-pregnancy BMI was calculated based upon self-reported pre-pregnancy weight and current height as measured during the first study-related clinic visit. Method of feeding at 1 month of age was based upon maternal self-report.
All testing took place in the University of Oklahoma Children's Physicians Building located at the University of Oklahoma Health Sciences Center. The institutional review board approved all procedures for human participants at the University of Oklahoma Health Sciences. Before testing all mothers signed an informed consent and a Health Insurance Portability and Accountability Act authorization.
Infant weight and length
For each visit, weight and length were obtained. A nude weight was obtained in duplicate using a Seca 728 electronic infant scale (Seca, Hamburg, Germany; accuracy ± 0.1 g). If both trials were within 10 g the results were averaged. Length was obtained in duplicate using a Seca 416 infantometer (Seca) with both trials having to be within 0.5 cm. In the instance that the weight and length measures were outside of the specified limits, a third measure was performed with the two closest values averaged. The same research assistant performed all measures of weight and length. Rate of weight gain at each distinct interval was calculated as the change in weight (in grams) divided by the number of days between measurements.
At 12 months of age, whole body composition (total fat and lean mass) and regional (trunk, arm, and leg) composition were determined by DXA using a Lunar iDXAv11-30.062 (Infant whole body analysis enCore 2007 software; GE, Fairfield, CT) scanner. During scanning, infants lay supine, wearing only a disposable diaper and swaddled in a light cotton blanket provided by the laboratory. If excessive movement occurred, the scan would be deemed invalid, and infants were not to be re-scanned. However, in this cohort, none of the infants moved excessively. The same person (D.A.F.) positioned all infants and analyzed all scans.
Of the 51 children that met the inclusion criteria for this study, three were excluded for the presence of maternal diabetes during pregnancy, and one was excluded because his rate of weight gain during the first year of life was deemed to be an outlier (i.e., >3 s.d. from the mean). Final analyses were conducted on the remaining 47 children. Simple Pearson correlations were calculated to examine the associations among maternal characteristics, BWLz, rates of weight gain during discrete periods of infancy, and body composition at 12 months. Independent groups' t-tests and ANOVA with Bonferroni post-hoc tests were used to examine whether body composition at 12 months varied by gender, mode of infant feeding, and Institute of Medicine (IOM) GWG categories. Stepwise linear regression modeling was used to further explore predictors of total lean and total fat mass, and of trunk fat mass. Variables selected for inclusion in each of the models included those with a simple correlation coefficient of at least 0.25 for the association with the dependent variable, or, in the case of categorical variables, a trend (i.e., P ≤ 0.10) for a between group difference in the outcome variable was observed. All analyses were conducted using SPSS software version 18.0 (SPSS, Chicago, IL). The α-level was set at P < 0.05 for statistical significance.
Descriptive statistics of the sample are shown in Table 1. Slightly more than half of the infants were female, whereas the majority of the sample was self-identified as white (81%). Fifty-seven percent of infants were exclusively breast-fed to 1 month of age, 43% exclusively formula-fed, whereas two infants received both breast milk and formula. Sixty percent of the mothers had a reported BMI in the normal weight range before pregnancy, 19% were overweight, and 17% were obese. Only two mothers gained inadequate weight during pregnancy according to the IOM criteria, with 49% meeting and 47% exceeding the 2009 IOM recommendations. This was the first pregnancy for 45% of the mothers, the second pregnancy for 28%, and the third or fourth pregnancy for the remaining 27%. Seventy percent of the infants were delivered between 39 and 41 weeks, with 28% delivering before 39 weeks and only one mother delivered beyond 41 weeks gestation.
Simple correlations among maternal and infant characteristics, infant rate of weight gain at discrete periods of infancy, and infant body composition at 12 months are shown in Table 2. None of the maternal characteristics was associated with infant BWLz. Older maternal age was associated with greater trunk and total fat mass at 12 months of age. There was a trend for maternal pre-pregnancy BMI to be positively associated with children's lean mass at 12 months (0.05 < P < 0.10). A greater rate of weight gain across the first year was associated with children's total lean mass, total fat mass, and trunk fat mass at 12 months. Gestational age at delivery and parity were not associated with any outcome of interest (data not shown).
Independent groups t-tests revealed that male infants gained weight more rapidly from 0-3 months (28.8 ± 6.4 vs. 24.7 ± 5.2 g/day; t = 2.289, P < 0.05), and had greater lean mass at 12 months of age (6.9 ± 0.6 vs. 6.3 ± 0.9 kg; t = 2.370, P < 0.05). There was no gender difference in BWLz or in total or trunk fat mass at 12 months. Formula-fed infants gained weight more rapidly from 6 to 12 months (12.4 ± 1.9 vs. 10.5 ± 2.9 g/day; t = 2.657, P < 0.05), but there was no difference in body composition at 12 months for infants that were formula-fed vs. breast-fed. With respect to maternal IOM category for GWG, rate of weight gain from 0-3 months was greater among infants of women who exceeded the recommendations (28.8 ± 5.0 vs. 24.6 ± 6.1 and 19.6 ± 4.8 g/day for infants of mothers who exceeded, met, and did not meet IOM recommendations, respectively; F = 4.577, P < 0.05), although post-hoc tests found only trends for the between groups comparisons (0.05 < P < 0.10). There was also a trend for greater trunk fat mass at 12 months among the infants of mothers who exceeded the recommendations (0.87 ± 0.25 vs. 0.81 ± 0.20 and 0.50 ± 0.03 kg for infants of mothers who exceeded, met, and did not meet IOM recommendations, respectively; F = 2.497, P = 0.094), however this effect was mainly attributable to the difference in trunk fat among infants of women who exceeded vs. did not meet the recommendations (P = 0.099).
Stepwise linear regression modeling was used to identify which of the maternal and child variables were independent predictors of child total lean mass, total fat mass, and trunk fat mass at 12 months. Results of the final models predicting body composition at 12 months are shown in Table 3. Rate of weight gain from 3-6 months, 6-12 months, and 0-3 months contributed 10% (partial r = 0.360, P < 0.05), 3.4% (partial r = 0.356, P < 0.05), and 3% (partial r = 0.386, P < 0.05), respectively, to total lean mass after adjusting for length (Table 3, model 1). Maternal pre-pregnancy BMI contributed a further 3% of the variance in children lean mass (partial r = 0.322, P < 0.05). Gender and maternal IOM category for GWG were not independent predictors of total lean mass at 12 months and so were excluded from the final model.
The second model in Table 3 shows the final results for the model predicting children's total fat mass at 12 months of age. The final model explained 71% of the variance in children's total fat mass. Children's rate of weight gain from 3-6 months explained 48% of the variance in children's total fat mass (partial r = 0.463, P < 0.01). Rate of weight gain from 0-3 months and 6-12 months contributed an additional 12% (partial r = 0.553, P < 0.001) and 7.7% (partial r = 0.515, P < 0.001) of the variance, respectively. Maternal age at delivery contributed 4% of the variance in children's total fat mass (partial r = 0.351, P < 0.05). Gender and total lean mass were excluded from the final model.
The final model predicting 63% of the variance in children's trunk fat mass is also shown in Table 3 (model 3). After adjusting for children's leg fat mass, 7.8 and 3.7% of the variability, respectively, was attributable to infant rate of weight gain at 0-3 months (partial r = 0.414, P < 0.01) and 3-6 months (partial r = 0.304, P < 0.05). Gender, maternal age at delivery, IOM category for GWG, and rate of weight gain from 6-12 months were not independent predictors of relative trunk fat at 12 months of age.
The objective of this study was to identify predictors of children's body composition at 1 year of age. Results showed that total lean mass, adjusted for length, was independently predicted by maternal pre-pregnancy BMI and children's rate of weight gain across the first year of life. Total fat mass at 1 year was predicted by maternal age at delivery and rate of weight gain across the first year of life. Trunk fat mass, relative to peripheral fat, was predicted by rate of weight gain during the first 6 months of life. These findings expand previous research by identifying specific risk factors and critical periods that may be important to body composition during early childhood.
In this cohort, lean mass at 1 year of age as measured by DXA, was predicted by more rapid weight gain across the first year, with that from 3-6 months making the largest contribution (10% of the variance). Few previous studies have examined and reported lean mass during early childhood. Chomtho et al. showed that more rapid weight gain from 0-12 weeks predicted lean mass at 12 weeks, whereas birth weight was not predictive of lean mass (17). Other studies have found that more rapid weight gain during the first 6 months of life was associated with fat-free mass in later childhood and adulthood (10,14,18). Maternal pre-pregnancy BMI also contributed a small but significant amount of variation in children's lean mass at 1 year of age. This finding is consistent with results of previous studies in which maternal pre-pregnancy weight or BMI was associated with offspring fat-free mass at birth (19), and during later childhood (20,21). Given the consistency of findings from different cohorts across childhood, it is possible that the association between maternal pre-pregnancy weight and offspring lean mass are attributable to shared genotype or to a chronic effect of the intrauterine environment to program offspring lean mass. It would be of interest in future longitudinal research to tease apart the effects of genotype from intrauterine environment, and to identify cut-points for healthy rates of weight gain during infancy that will preserve the optimal development of lean mass.
In this cohort, children's total fat mass at 12 months was predicted by rate of weight gain across the first year of life. Weight gain velocity during the first 6 months had a particularly strong influence (62% of the variance in total fat mass), whereas rate of weight gain at 6-12 months contributed only 8%. This finding supports and extends results of studies among older children (9,11,18,22). Although the mechanisms that underlie the association between rate of weight gain during early infancy and subsequent adiposity have not been elucidated, there is good evidence in the literature to suggest that infant feeding practices play a significant role. For example, formula-fed infants have greater risk for obesity during childhood (23,24). Although we did not find an effect of formula-feeding on adiposity at 12 months in the current study, the fact that formula-fed infants gained weight more rapidly from 6-12 months is consistent with a potential obesogenic effect of formula-feeding. It is possible that with the small sample size, the current study may have been insufficiently powered to identify an effect of formula-feeding on adiposity at such a young age. In addition, because mode of feeding was captured only at 1 month of age in this study, it is possible that feeding mode in subsequent months may have been associated with adiposity at 12 months. Other infant feeding practices that were not measured in this study, such as early (i.e., ≤4 months) introduction of solid food (25-,26, 27), and scheduled rather than infant-led feedings (23), may also contribute to the strong influence of weight gain velocity during the first 6 months of life on adiposity at 12 months. Together with the existing literature, therefore, these results highlight the possibility that interventions to improve infant feeding practices before 6 months may prove efficacious in reducing subsequent adiposity.
Offspring body fat at 1 year of age was also associated with maternal age at delivery. Although we cannot determine from this study, what aspect of age was responsible for this association, there are two likely possibilities. First, it is possible that age is a proxy for parity, and indeed, maternal age at delivery was positively associated with parity in this cohort (r = 0.333, P < 0.05). In previous studies, a positive association between parity and offspring body fat at birth has been observed (28,29). However, in this cohort, parity was not associated with offspring BWLz or body fat at 1 year of age, possibly due to the small sample size. The effect of maternal age on offspring adiposity might also have been attributable to age-related differences in metabolic health during pregnancy. For example, glucose tolerance declines with age in the general population (30), and this effect likely contributes to the fact that advancing maternal age is associated with greater risk for gestational diabetes (31,32,33). Lower glucose tolerance and consequently, higher circulating glucose concentrations, even among nondiabetic mothers, is believed to influence developmental programming of offspring adiposity (34,35). Consequently, the modest but significant effect of maternal age on offspring adiposity may have reflected either greater parity or subtle age-associated impairments of metabolic health.
We also examined whether trunk fat, a measure of central fat, was associated with gestational characteristics and infant rate of weight gain. Central fat is associated with adverse metabolic health in both adults and children (15,16). We found that the proportion of fat deposited to the trunk vs. the periphery was predicted by rate of weight gain during the first 6 months of life. Although the effects were modest (7.8 and 3.7% of the variance attributed to rate of weight gain from 0-3 months and 3-6 months, respectively), it is interesting that only weight gain during the first 6 months, but not the second 6 months of life contributed to relative trunk fat mass. Similarly in another study of healthy infants, weight gain during the first 6 months of life was associated with trunk fat at 6 months (36). Research among small-for-gestational-age infants likewise suggests that rapid weight gain during infancy leads to continued increase in abdominal obesity beyond infancy (37). Taken together, these observations allude to a potential effect of early infancy weight gain on the programming of future patterns of fat deposition. Future longitudinal studies are needed, however, to examine whether abdominal obesity at 12 months of age persists until adulthood.
Strengths of this study include the longitudinal assessment of children's growth during infancy and the use of DXA to measure total and regional body composition at 1 year of age. This study was also strengthened by the concurrent consideration of variables before and following birth that may have impacted children's body composition. This study was limited by the small sample size, reliance upon self-report for maternal pre-pregnancy weight and GWG, the lack of information regarding feeding mode beyond 1 month of age, and by the relatively homogeneous nature of the sample, which limits generalizability to more diverse populations.
In conclusion, this study extends the existing literature by showing independent contributions of maternal variables and infant weight gain velocity during the first year of life with subsequent body composition and fat distribution at 1 year. Although we cannot determine, on the basis of these data, whether adiposity at 1 year will track across childhood; evidence from longitudinal studies using simple measures of BMI and weight-for-length supports this possibility (38-,39,40). The data presented here highlight the importance of examining influences on growth during early infancy, and to identify factors that contribute to unhealthy patterns of fat distribution. Furthermore, the first 6 months of life may provide a critical opportunity during which interventions could impact future metabolic health.
The authors acknowledge Brittney Criswell and Catherine Wolf for coordinating study visits and the collection and entry of data. This study was partially funded by Nestlé.