What hormones are involved in growth and development during infancy and childhood?

Journal Article

Anni Larnkjær,

1From the Department of Human Nutrition Centre for Advanced Food Studies Faculty of Life Sciences Frederiksberg C Denmark (AL LS-N CM HKIKFM)the Department of Biomedical Sciences Panum Institute (JJH) University of Copenhagen Denmark.

3Address correspondence to A Larnkjær, Department of Human Nutrition, Faculty of Life Sciences, University of Copenhagen, Rolighedsvej 30, DK-1958 Frederiksberg C, Denmark. E-mail: .

Search for other works by this author on:

• 
  PDF
• Split View
• • Article contents
  • Figures & tables
  • Video
  • Audio
  • Supplementary Data

• Cite

  Cite

  Anni Larnkjær, Lene Schack-Nielsen, Christian Mølgaard, Helga K Ingstrup, Jens J Holst, Kim F Michaelsen, Effect of growth in infancy on body composition, insulin resistance, and concentration of appetite hormones in adolescence, The American Journal of Clinical Nutrition, Volume 91, Issue 6, June 2010, Pages 1675–1683, https://doi.org/10.3945/ajcn.2009.27956

  Close

• • Email
  • Twitter
  • Facebook
  • More

Close

Navbar Search Filter Microsite Search Term Search

ABSTRACT

INTRODUCTION

The global epidemic of childhood obesity is a major public health issue because most obese children remain obese as adults. Secular changes in eating behavior and genetic susceptibility are factors that have received much attention. However, early programming has been suggested as a potential important mechanism that could contribute to obesity and insulin resistance (IR). High birth weight (BW) has been associated with a greater propensity toward obesity later in life and a low BW with IR and increased risk of diabetes mellitus, which suggests that fetal life is a critical window for programming of later disease risks (1–3). There is also increasing evidence that postnatal programming play a role in body composition and IR. Both diet and weight gain during the first years of life are associated with a later risk of obesity and IR (4–6). Some studies have shown breastfeeding to be protective against childhood and adult obesity (7, 8), and breastfeeding is associated with a slower growth in infancy (9, 10). A systematic review from 2003 (5) concluded that infants in the high end of the distribution for weight or body mass index (BMI), or with a rapid weight gain during infancy, are at increased risk of subsequent obesity. It was not possible to identify a period during the first 2 y of age that was more critical than other periods. Furthermore, an interaction with BW was not shown in any of the studies (ie, high infancy weight gain was a risk factor for obesity in later life irrespective of BW). Many studies used BMI to classify obesity, but measurements reflecting adipose tissue such as dual-energy X-ray absorptiometry (DXA) scanning may better reflect later health consequences. In addition, most studies analyzed the effect of one period during infancy with only one follow-up visit.

The mechanisms behind a potential association between rapid weight gain during infancy and the propensity for obesity in later life are not well established. It was suggested that altered function of central regulatory mechanisms may affect later obesity, possibly through programming of the appetite regulatory hormones (11, 12). Obesity is associated with a high concentration of leptin (13) and low concentrations of ghrelin (14, 15) and adiponectin (16, 17). The influence of weight gain in infancy on later concentrations of these appetite regulatory hormones is not known. One study(18) suggested that early diet may program later adiposity-corrected leptin concentrations, but to our knowledge, it has not been investigated if accelerated weight gain in infancy is associated with later concentrations of leptin, ghrelin, and adiponectin; if this is the case, one explanation could be that the concentrations of these hormones are programmed early in life.

The aim of the current study was to examine if weight gain during the first 9 mo of life was associated with body-composition measures at the ages of 10 and 17 y and with IR and absolute or adiposity-corrected concentrations of leptin, ghrelin, and adiponectin in adolescence in a cohort of healthy term infants with a BW appropriate-for-gestational-age (AGA).

SUBJECTS AND METHODS

This study was part of the Copenhagen Cohort Study on Infant Nutrition and Growth, a prospective observational study in which a random sample of Danish infants born from 1987 to 1988 was followed during the first year of life. The inclusion criteria were as follows: parents of Danish origin, a singleton birth, gestational age between 37 and 42 wk, BW for gestational age between 10th and 90th percentiles, and no neonatal disease or malformation. Of the 251 infants fulfilling the inclusion criteria, 142 infants completed the examination during infancy. Of these, 59 subjects constituted a control group that was only examined at 9 mo. The other subjects were examined several times during the first year of life. A more detailed description of the selection and characteristics of the study group was published previously (19).

Follow-up

Participants from the original study were invited to participate in the follow-up studies when they were 10 and 17 y old. At the follow-up at 10 and 17 y of age, 105 and 109 subjects, respectively, agreed to participate. In this analysis, 95 subjects who completed the 17-y examination and with full information on the covariates were included in the analysis. Eighty-two subjects (86%) of the 95 children also participated in the 10-y follow-up examination. The study was approved by the Ethics Committee for Copenhagen and Frederiksberg (KF 01–226/97).

Examination during first year of life

Weight was measured with the infant naked on an electronic balance to the nearest gram. Weight measurements were available for a subgroup of 60 subjects at 3 and 6 mo of age. Recumbent length was measured on a wooden measuring board with a fixed headboard and a moveable foot. Birth data were achieved from birth records. The duration of breastfeeding was registered, and the infants were classified as breastfed if they were breastfed at least once a day.

Follow-up examinations at 10 and 17 y of age

Height was measured to the nearest 1 mm by a wall-mounted stadiometer (Chasmore, London, United Kingdom), and weight was measured to the nearest 0.1 kg by an electronic scale (Lindeltronic 8000; Lindells Inc, Kristianstad, Sweden). Triceps skinfold thickness, measured with a Harpenden Skinfold Caliper (Chasmors Ltd, London, United Kingdom), and waist and hip circumference were measured in triplicate according to standard procedures, and the measures were averaged.

Body fat (BF), percentage BF, percentage of trunk fat relative to whole BF (percentage TF/BF), and percentage trunk fat (percentage TF) were measured on a Hologic quantitative digital radiography 1000/W DXA-scanner (Hologic Inc, Waltham, MA) by using software versions 5.61 and 5.73 at the 10- and 17-y follow-up examinations, respectively. Subjects wore only underpants and a cotton T-shirt during the scan. The entrance radiation dose amount was 15 μSv with an effective dose ≤10 μSv, which is equal to a background radiation of ≈1 d in Denmark. The TF was assessed by subtraction of fat mass from the head, arms, and legs from the total BF mass.

Hormone and biochemical assays

At the 17-y follow-up examination, overnight fasting blood samples were drawn from vein puncture between 0800 and 1000. Plasma was stored at −80°C until analysis. Plasma concentrations of leptin, total ghrelin, and adiponectin were measured by commercially available radioimmunoassay kits (Linco Research Inc, St Charles, MO). The intra- and interassay CVs were, respectively, 6% and 8% for leptin, 10% and 15% for ghrelin, and 2% and 9% for adiponectin. Homeostasis model assessment (HOMA) was used to estimate IR, which was calculated as follows: [fasting insulin (μU/mL) × fasting glucose (mmol/L)/22.5]. Insulin concentrations were measured by chemiluminescence immunologic assay (IMMULITE/IMMULITE 1000 Insulin, DPC; Diagnostic Products Corp, Los Angeles, CA) with intra- and interassay CVs of 1.8% and 6.2%, respectively. Glucose was measured by the hexokinase method on a Cobas Mira automated biochemistry analyzer (Hoffmann-La Roche & Co, Basel, Switzerland) with kits (Roche Diagnostics, Mannheim, Germany). The intra- and interassay CVs were 1.0% and 1.4%, respectively.

Calculations

To adjust for the exact age at measurement and sex differences, sex- and age-independent SD scores for weight (weight-SDSs) at birth and 3, 6, and 9 mo of age were calculated by using World Health Organization reference-data growth charts (20), and the change in weight-SDSs (Δweight-SDSs) from birth to 9 mo of age was calculated. For the subgroup, Δweight-SDS was also calculated for the periods from birth to 3 mo of age, 3 to 6 mo of age, and 6 to 9 mo of age.

Statistical analyses

Results are shown as means ± SDs for descriptive variables. P ≤ 0.05 was considered statistically significant. Differences in the variables between girls and boys and between the study group and the subgroup were tested by the independent t test or, in the case of nonparametric variables, by the Mann-Whitney test.

The effects of weight gain in infancy on body composition (ie, BMI, percentage BF, percentage TF/BF, percentage TF, triceps skinfold thickness, and waist-hip ratio), HOMA, and appetite regulatory hormones (leptin, ghrelin, and adiponectin) as absolute values and corrected for adiposity (both as concentration/BF and concentration/percentage BF) were investigated by multiple linear regression by using general linear models. Residual plots and Cook’s distance were used for model verification and identification of outliers. Nonlinearity of the regression models was tested by including a quadratic term. The appetite regulatory hormone variables were log transformed for linear regression analyses. Therefore, a coefficient of, eg, 0.86 in the regression analyses for the appetite regulatory hormones corresponds to a 14% decrease in hormone concentration per Δweight-SDS unit.

We evaluated the effect of weight gain from 0 to 9 mo of age in different models with 10-y and 17-y values. The first model was adjusted for sex, and the second model was adjusted for sex and covariates. A third model used in the analyses for body-composition variables at 17 y of age was additionally adjusted for the measure of body composition in question at 10 y of age. The effect of Δweight-SDS during the first 3 mo of life and from 3 to 6 mo of age, 6–9 mo of age, and birth to 9 mo of age on body composition, IR, appetite hormones, and adiposity-corrected appetite hormones was tested in the subgroup at 17 y of age in the first 2 models. Initial analyses indicated that there was no significant interaction between sex and Δweight-SDS. Parental BMI was significantly associated with BMI at 17 y of age (BMI of mothers: r = 0.223, P = 0.027; BMI of fathers: r = 0.228; P = 0.026) and leptin (BMI of mothers: r = 0.369, P < 0.001; BMI of fathers: r = 0.226, P = 0.028) and, therefore, was included as a covariate. BW was significantly negatively correlated with Δweight-SDS from 0 to 9 mo of age (r = −0.512, P ≤ 0.001). Preliminary analysis indicated that duration of breastfeeding was not likely to be a confounder and, therefore, was not included in the analyses. Social class was also not included as a covariate because it was not significantly associated with Δweight-SDS from 0 to 9 mo of age (P = 0.283), BMI at 17 y of age (P = 0.339), or percentage BF (P = 0.864). Correction for parental height and weight in addition to the parental BMI in preliminary analyses had no influence on the results. Furthermore, no independent effect of height was shown in the preliminary analysis. The study was powered to show a difference of ≈0.57 × SD in the outcome variables with a power of 80 for the core group. Statistical analyses were performed with SPSS (version 15.0; SPSS Inc, Chicago, IL).

RESULTS

The characteristics of the subjects are presented in Table 1. Boys were heavier than girls at birth and at 9 mo of age, whereas there was no significant difference between boys and girls regarding the increase in weight-SDSs from 0 to 9 mo of age. At both 10 and 17 y of age, the measures of fat in the body (percentage BF, percentage TF/BF, percentage TF, waist-hip ratio, and triceps skinfold thickness) were higher for girls than for boys, whereas there was no sex difference for BMI. According to international cutoffs (21), 5 children (2 boys) were overweight, and none of the children were obese, at the 10-y examination, whereas 8 (4 boys) adolescents were overweight, and 2 (1 boy) adolescents were obese, at the 17-y follow-up visit. For parents, 1 mother was obese, and 10 mothers were overweight, whereas 4 fathers were obese, and 23 fathers were overweight. Of the 32 children with ≥1 overweight or obese parent, 4 (1 boy) children were overweight, and 2 (1 boy) children were obese, at 17 y of age.

TABLE 1

Infant, childhood, and adolescent characteristics of the subjects1

1

ΔWeight-SDS, change in SD scores for weight.

2

Mean ± SD (all such values).

3

Significantly different from boys, P ≤ 0.01 (independent t test or Mann-Whitney test).

TABLE 1

Infant, childhood, and adolescent characteristics of the subjects1

1

ΔWeight-SDS, change in SD scores for weight.

2

Mean ± SD (all such values).

3

Significantly different from boys, P ≤ 0.01 (independent t test or Mann-Whitney test).

Overall, 23% of the children displayed an increase in weight-SDSs > 0.67 from birth to 9 mo of age. This value has been proposed as the limit for defining rapid weight gain and corresponds to the difference between the 25th and 50th percentiles or the 50th and 75th percentiles in a growth chart (22, 23). All infants had Δweight-SDSs > −2 except for one infant with a Δweight-SDS = −2.09, and 74% of the infants had Δweight–SDSs between −1 and 1 (Figure 1). At birth, all children had weight-SDSs higher than the −2.0 threshold and lower than the 2.0 threshold because of the inclusion criteria.

FIGURE 1

Distribution of change in SD scores for weight (weight-SDS) from 0 to 9 mo of age in 95 appropriate-for-gestational-age term infants.

The effect of Δweight-SDS during the first 9 mo of life on measures for body composition at 17 y of age were tested in linear regression models adjusted for sex with and without confounders and with Δweight-SDS from 0 to 9 mo of age as a continuous variable (Table 2). Weight gain during the first 9 mo of life was positively related to BMI, percentage BF, percentage TF, and triceps skinfold thickness with adjustment for sex or sex and covariates. ΔWeight-SDS from 0 to 9 mo of age was positively correlated with the waist-hip ratio controlled for sex, but when both sex and covariates were included in the analysis, there was only a trend. Weight gain in infancy was not associated with percentage TF/BF in late adolescence. The partial r values were higher for BMI than for the variables describing the fat content of the body, suggesting that there might also be an effect of early weight gain on later body size. There was a significant effect of sex regarding percentage BF, percentage TF/BF, percentage TF, and triceps skinfold thickness corresponding to higher values of 12%, 3%, 10%, and 7 mm, respectively, for girls.

TABLE 2

Multiple linear regression analysis of the association between change in SD scores for weight from birth to 9 mo of age and measures of body composition at 17 y of age

1

General linear model adjusted for sex.

2

General linear model adjusted for sex, birth weight, father’s BMI, and mother’s BMI.

3

Coefficients in regression (B) and SE.

4

Significant covariates in the analysis: BMI, father’s BMI (P = 0.024); body fat, sex (models 1 and 2, P ≤ 0.001), and father’s BMI (P = 0.029); trunk fat/body fat, sex (model 1, P = 0.004; model 2, P = 0.020); trunk fat, sex (P ≤ 0.001); and triceps skinfold thickness, sex (P ≤ 0.001).

5

Subgroup with anthropometric data at 3 and 6 mo of age.

6

Significant covariates in the analysis: BMI, father’s BMI (P = 0.025); body fat, sex (P ≤ 0.001); trunk fat/body fat, sex (model 1, P = 0.024); trunk fat, sex (P ≤ 0.001); waist-hip ratio, father’s BMI (P = 0.034); and triceps skinfold thickness, sex (P ≤ 0.001).

TABLE 2

Multiple linear regression analysis of the association between change in SD scores for weight from birth to 9 mo of age and measures of body composition at 17 y of age

1

General linear model adjusted for sex.

2

General linear model adjusted for sex, birth weight, father’s BMI, and mother’s BMI.

3

Coefficients in regression (B) and SE.

4

Significant covariates in the analysis: BMI, father’s BMI (P = 0.024); body fat, sex (models 1 and 2, P ≤ 0.001), and father’s BMI (P = 0.029); trunk fat/body fat, sex (model 1, P = 0.004; model 2, P = 0.020); trunk fat, sex (P ≤ 0.001); and triceps skinfold thickness, sex (P ≤ 0.001).

5

Subgroup with anthropometric data at 3 and 6 mo of age.

6

Significant covariates in the analysis: BMI, father’s BMI (P = 0.025); body fat, sex (P ≤ 0.001); trunk fat/body fat, sex (model 1, P = 0.024); trunk fat, sex (P ≤ 0.001); waist-hip ratio, father’s BMI (P = 0.034); and triceps skinfold thickness, sex (P ≤ 0.001).

When the anthropometric measures at 10 y of age were included as covariates in the model, the weight gain in infancy was no longer significant for any of the variables. Instead the 10-y values were highly significant (P < 0.001) for all variables except for waist-hip ratio (P = 0.136) (results not shown). Therefore, we also analyzed the effect of early weight gain on body composition at 10 y of age. The results were consistent with the findings in late adolescence except for the waist-hip ratio, which at 10 y of age was not significantly associated with weight gain in infancy in the model adjusted for sex. Otherwise, the same significant associations between Δweight-SDS from 0 to 9 mo of age and body composition were present at 10 and 17 y of age. The effect for every 1.0 SD from weight gain from 0 to 9 mo of age was ≈0.3 BMI SD units corresponding to 1.4 and 1.7 kg at 10 and 17 y of age, respectively, using a mean height. There was a significant effect of sex on percentage BF, percentage TF/BF, percentage TF, triceps skinfold thickness, and waist-hip ratio, with higher values for girls, except for the waist-hip ratio, corresponding to 4%, 9%, 4%, and 2 mm for higher percentage BF, percentage TF/BF, percentage TF, and triceps skinfold thickness, respectively, in girls (results not shown). The quadratic term (Δweight-SDS)2 included in the models to control for nonlinearity was not significant in any of the models.

For a subgroup of 60 participants, weight measurements were available at 3 and 6 mo of age, and Δweight-SDSs from birth to 3 mo of age, 3 to 6 mo of age, and 6 to 9 mo of age were calculated for this subgroup. The subgroup was representative of the whole sample because there were no significant differences in the characteristics or the parental characteristics between groups (data not shown). The effects of Δweight-SDS from birth to 9 mo of age on body-composition variables in late adolescence in the subgroup showed the same pattern as in the whole group (Table 2).

We tested if the effect of Δweight-SDS during the first 9 mo of age on the body composition measures at 17 y of age could be confined to the Δweight-SDS in one of the three 3-mo periods during the first 9 mo of life. Analyses revealed that the effect on later body composition mainly arose from the Δweight-SDS during the first 3 mo of life, as there were no significant associations between Δweight-SDSs from 3 to 6 or 6 to 9 mo of age and body-composition variables at 17 y of age. The results for the 0–3-mo period are shown in Table 3, whereas the data for the 3–6- and 6–9-mo periods are not shown.

TABLE 3

Multiple linear regression analysis of the association between change in SD scores for weight from birth to 3 mo of age and measures of body composition at 17 y of age in the subgroup (n = 60)

1

Significant covariates in the analysis: body fat, sex (models 1 and 2, P ≤ 0.001), and father’s BMI (P = 0.046); trunk fat/body fat, sex (model 1, P = 0.023); trunk fat, sex (model 1 and 2, P ≤ 0.001), and father’s BMI (P = 0.037); waist-hip ratio, father’s BMI (P = 0.018); and triceps skinfold thickness, sex (model 1, P = 0.001; model 2, P = 0.002).

2

General linear model adjusted for sex.

3

General linear model adjusted for sex, birth weight, father’s BMI, and mother’s BMI.

4

Coefficients in regression (B) and SE.

TABLE 3

Multiple linear regression analysis of the association between change in SD scores for weight from birth to 3 mo of age and measures of body composition at 17 y of age in the subgroup (n = 60)

1

Significant covariates in the analysis: body fat, sex (models 1 and 2, P ≤ 0.001), and father’s BMI (P = 0.046); trunk fat/body fat, sex (model 1, P = 0.023); trunk fat, sex (model 1 and 2, P ≤ 0.001), and father’s BMI (P = 0.037); waist-hip ratio, father’s BMI (P = 0.018); and triceps skinfold thickness, sex (model 1, P = 0.001; model 2, P = 0.002).

2

General linear model adjusted for sex.

3

General linear model adjusted for sex, birth weight, father’s BMI, and mother’s BMI.

4

Coefficients in regression (B) and SE.

Appetite regulatory hormones

The absolute plasma concentrations of leptin, ghrelin, and adiponectin and adiposity-corrected concentrations expressed relative to BF and percentage BF are presented in Table 4. Compared with boys, girls had significantly higher concentrations of leptin, leptin corrected for adiposity, ghrelin, and adiponectin and lower concentrations of ghrelin and adiponectin, both corrected for adiposity. Leptin was positively and ghrelin was negatively associated with all body composition variables at 17 y of age and controlled for sex with r = 0.83 to 0.46 for percentage BF and waist-hip ratio for leptin and r = −0.42 to −0.23 for BMI and waist-hip ratio, respectively, for ghrelin. However, there were no associations between adiponectin and body-composition measures adjusted for sex.

TABLE 4

Physiologic characteristics of subjects at 17 y of age (n = 95)1

1

Values are means ± SDs. BF, body fat; HOMA-IR, homeostasis model assessment for insulin resistance.

2

Significantly different from boys, P ≤ 0.005 (independent t test or Mann-Whitney test).

TABLE 4

Physiologic characteristics of subjects at 17 y of age (n = 95)1

1

Values are means ± SDs. BF, body fat; HOMA-IR, homeostasis model assessment for insulin resistance.

2

Significantly different from boys, P ≤ 0.005 (independent t test or Mann-Whitney test).

The weight gain during the first 9 mo of life was not significantly related to later concentrations of leptin (Table 5). ΔWeight-SDS in infancy was significantly negatively related to ghrelin corrected for adiposity: in adolescence, an increase of 1 weight-SDS from 0 to 9 mo of age would reduce the concentrations of ghrelin per BF and ghrelin per percentage BF by 14% and 11%, respectively, after controlled for sex. The associations remained significant after adjustment for other covariates (sex, BW, and parental BMI) with an increase in R2 from 0.15 to 0.27 for ghrelin per kilogram BF and from 0.30 to 0.33 for ghrelin per percentage BF compared with the basic model. There was a significant effect of sex corresponding to 1.4- and 1.7-times higher concentrations of ghrelin per BF and ghrelin per percentage BF for boys, respectively.

TABLE 5

Multiple linear regression analysis of the association between change in SD scores for weight from birth to 9 mo of age and appetite regulatory hormones at 17 y of age1

1

BF, body fat.

2

General linear model adjusted for sex.

3

General linear model adjusted for sex, birth weight, father’s BMI, and mother’s BMI.

4

Coefficients in regression (B) and 95% CIs after back transformation of log-transformed values.

5

Significant covariates in the analysis—leptin, leptin/BF, and leptin/percentage BF: sex (P ≤ 0.001) and mother’s BMI (P ≤ 0.01); ghrelin, ghrelin/BF, and ghrelin/percentage BF: sex (P ≤ 0.01); adiponectin, adiponectin/BF, and adiponectin/percentage BF: sex (P ≤ 0.05); and adiponectin/BF: father’s BMI (P = 0.013).

6

Subgroup with anthropometric data at 3 and 6 mo of age.

7

Significant covariates in the analysis—leptin, leptin/BF, and leptin/percentage BF: sex (P ≤ 0.001); ghrelin and ghrelin/percentage BF: sex (P ≤ 0.05); adiponectin: sex (P ≤ 0.001); adiponectin/BF: father’s BMI (P = 0.026); and adiponectin/percentage BF: sex (P ≤ 0.05).

TABLE 5

Multiple linear regression analysis of the association between change in SD scores for weight from birth to 9 mo of age and appetite regulatory hormones at 17 y of age1

1

BF, body fat.

2

General linear model adjusted for sex.

3

General linear model adjusted for sex, birth weight, father’s BMI, and mother’s BMI.

4

Coefficients in regression (B) and 95% CIs after back transformation of log-transformed values.

5

Significant covariates in the analysis—leptin, leptin/BF, and leptin/percentage BF: sex (P ≤ 0.001) and mother’s BMI (P ≤ 0.01); ghrelin, ghrelin/BF, and ghrelin/percentage BF: sex (P ≤ 0.01); adiponectin, adiponectin/BF, and adiponectin/percentage BF: sex (P ≤ 0.05); and adiponectin/BF: father’s BMI (P = 0.013).

6

Subgroup with anthropometric data at 3 and 6 mo of age.

7

Significant covariates in the analysis—leptin, leptin/BF, and leptin/percentage BF: sex (P ≤ 0.001); ghrelin and ghrelin/percentage BF: sex (P ≤ 0.05); adiponectin: sex (P ≤ 0.001); adiponectin/BF: father’s BMI (P = 0.026); and adiponectin/percentage BF: sex (P ≤ 0.05).

The concentration of adiponectin relative to BF adjusted for sex was also significantly negatively correlated with weight gain in infancy corresponding to a decrease in adiponectin per BF of 11% per unit increase in weight-SDS. The association of adiponectin corrected for BF to weight gain during the first 9 mo of life remained significant after adjustment for covariates. These variables accounted for 15% of the variance in adiponectin per BF in late adolescence, which was an increase of 7% compared with the basic model. Adiponectin and adiponectin corrected for percentage BF were not or tended not to be significantly associated with weight gain in infancy.

The effects of Δweight-SDS from birth to 9 mo of age on appetite hormone variables at the age of 17 y in the subgroup were examined. Overall, the pattern in the subgroup was the same as the pattern in the whole group (Table 5). The effect of Δweight-SDS in 3-mo intervals during the first 9 mo of life was also investigated for appetite hormones and appetite hormones corrected for adiposity in the subgroup. There were no significant differences in the concentrations of the appetite hormones between the subgroup and core group (data not shown). There were no significant associations between Δweight-SDS from 3 to 6 and 6 to 9 mo of age on any of the appetite hormone variables (data not shown), whereas several of the associations with the 0–3-mo-old group was significant (Table 6). ΔWeight-SDS from birth to 3 mo of age was significantly correlated with ghrelin per BF with and without adjustment for covariates. There was a trend (P = 0.056) for a significant association between Δweight-SDS from 0 to 3 mo of age and ghrelin relative to percentage BF, which became significant after adjustment for covariates. All adiponectin variables were significantly associated with Δweight-SDS during the first 3 mo of life, whereas none of the leptin variables correlated with weight gain from birth to 3 mo of age. The associations between weight gain from 0 to 3 mo of age and the concentrations of appetite hormones were stronger than for the 0–9-mo period for the subgroup.

TABLE 6

Multiple linear regression analysis of the association between change in SD scores for weight from birth to 3 mo of age and appetite regulatory hormones at 17 y of age in the subgroup (n = 60)1

1

BF, body fat.

2

Significant covariates in the analysis—leptin, leptin/BF, and leptin/percentage BF: sex (P ≤ 0.001); ghrelin and ghrelin/percentage BF: sex (P ≤ 0.05); adiponectin: sex (P ≤ 0.001); adiponectin/BF: father’s BMI (P = 0.026); and adiponectin/percentage BF: sex (P ≤ 0.05).

3

General linear model adjusted for sex.

4

General linear model adjusted for sex, birth weight, father’s BMI, and mother’s BMI.

5

Coefficients in regression (B) and 95% CIs after back transformation.

TABLE 6

Multiple linear regression analysis of the association between change in SD scores for weight from birth to 3 mo of age and appetite regulatory hormones at 17 y of age in the subgroup (n = 60)1

1

BF, body fat.

2

Significant covariates in the analysis—leptin, leptin/BF, and leptin/percentage BF: sex (P ≤ 0.001); ghrelin and ghrelin/percentage BF: sex (P ≤ 0.05); adiponectin: sex (P ≤ 0.001); adiponectin/BF: father’s BMI (P = 0.026); and adiponectin/percentage BF: sex (P ≤ 0.05).

3

General linear model adjusted for sex.

4

General linear model adjusted for sex, birth weight, father’s BMI, and mother’s BMI.

5

Coefficients in regression (B) and 95% CIs after back transformation.

IR

There was no significant association between weight gain during the first 9 mo of life and HOMA at 17 y of age adjusted for sex (P = 0.294) or adjusted for sex and covariates (P = 0.180). Control for current BMI in the models did not change the results. The effect of weight gain during the different 3-mo periods for the subgroup was analyzed, but there were no significant associations (data not shown). However, there was a significant negative correlation adjusted for sex between HOMA and adiponectin corrected for BF (r = −0.269, P = 0.009).

DISCUSSION

The current study of a cohort of term AGA infants showed that a rapid increase in weight-SDS during the first 9 mo of life was positively associated with several measures of BF (BMI, percentage BF, percentage TF, and triceps skinfold thickness) in late adolescence. A weight acceleration equal to +1 SDS from 0 to 9 mo of age was associated with an increase in BMI and percentage BF of ≈ +0.3 SD. The associations were already present at the age of 10 y, which indicated that the effects of Δweight-SDS on body composition persisted from childhood into late adolescence. Interestingly, the initial weight gain did not seem to be associated with central adiposity measured as percentage TF/BF. Likewise, we did not find any effect on IR or leptin. However, an increase in weight-SDS during infancy, especially in the first 3 mo of life, was associated with lower concentrations of ghrelin and adiponectin corrected for BF.

Several studies (22, 24) linked rapid weight gain during infancy to childhood or adult overweight and obesity. Our study is different from most of these studies as only a few of the children in our study were overweight and we measured BF, and not only BMI as was measured in most other studies. Furthermore, all children in our study were born AGA, and hence, the associations we find are not likely to be caused by catch-up growth of children with low BW. Our findings are in agreement with a study (25) that showed that rapid weight gain between 0 and 2 y of AGA infants resulted in increased BMI and percentage BF calculated from skinfold thickness, but the outcomes were only investigated until the age of 7 y. We showed that the effects on BF seen in prepubertal children are also present in late adolescence. This is consistent with the findings in a large Swedish study (26) where weight gain from birth to 1 y of age was positively associated with BF mass on the basis of bioelectric impedance at age 15 y and a study (27) that showed that most overweight before puberty is already present at 5 y of age.

The presence of critical windows in infancy for programming later overweight has been hypothesized. The analyses of the subgroup suggested that Δweight-SDS from 0 to 3 mo of age is most critical for later BMI and percentage BF. This is in agreement with other studies (4, 5, 28) that suggested that the first weeks or months of life are most important for later BMI and BF. Botton et al (29) investigated the effect of growth velocities at different ages and reported significant effects of weight gain at both 3 and 6 mo of age on BMI and BF measured by bioelectric impedance in adolescents. Likewise Leunissen et al (30) showed that rapid weight gain from 0 to 3 mo of age was associated with increased BF measured by DXA in young adults.

We showed no significant association between Δweight-SDS in infancy and later IR. Rapid weight gain in infancy was reported to program later IR (6, 30–32). However, the subjects were often born preterm or small for gestational age (SGA) (6, 31, 32) or the proportion of SGA subjects were high (30). Flinken et al (33) reported that Δweight-SDS from 0 to 3 mo of age predicted higher IR in 19-y-olds born preterm, but the association was borderline significant. Ekelund et al (34) showed that Δweight-SDS from 0 to 3 mo of age was associated with clustered metabolic risk at 17 y of age but no independent effect of insulin or glucose in a cohort with few preterm subjects. These results indicate that the effect is most pronounced in preterm and SGA subjects and might not exist in term AGA infants as our data suggest.

Nevertheless, IR was negatively correlated with adiponectin corrected for BF, and early weight gain was negatively associated with adiponectin concentrations at 17 y of age. Several studies indicated that adiponectin may be programmable. BW was reported to be positively associated with adiponectin concentrations at 2 y of age (35), in later childhood, (16, 31) and in adulthood (36), but another study (37) showed no associations. Catch-up growth was negatively associated with adiponectin in children with a mean age of 8.6 y and born SGA (31). Adiponectin improves insulin metabolism, and low concentrations may predict future IR (31, 38). The lower concentrations of adiponectin adjusted for fatness in adolescents with rapid weight gain in infancy could indicate a higher risk of developing IR if adiponectin is a more sensitive or earlier marker for IR than HOMA in subjects born term and AGA because there was no independent effect on IR. However, this was only an observational study, and more studies in AGA subjects are needed to clarify this observation.

We showed no significant association between Δweight-SDS in infancy and leptin variables, which could be due to the small sample size and because only term infants were examined. Low BW was associated with higher leptin concentrations adjusted for BMI in adult life (13, 39). Another study (18) reported that adiposity-corrected leptin concentrations in 13–16 y-old adolescents born preterm were higher in children who were randomly assigned to receive a nutrient-enriched formula than children who received either a standard formula or banked breast milk in infancy, which suggested that leptin concentrations may be programmed by early diet.

Ghrelin concentration corrected for BF was associated negatively with Δweight-SDS in infancy. To our knowledge, the effect of nutrition and growth in infancy on later ghrelin concentrations has not been extensively studied. A rat study of brain slices of adult offspring subjected to maternal food restriction during pregnancy and/or lactation suggested that prenatal and early postnatal nutrient concentrations may program pathway responses of ghrelin (40). Besides being orexigenic, ghrelin may inhibit adipogenesis and stimulate growth hormone release and is associated with glucose metabolism (14). However, ghrelin was not shown to be an independent predictor of future weight gain in a prospective study of cardiovascular risk factors (15) or metabolic syndrome in children (41). Our results suggest a programming effect of rapid weight gain in infancy on later concentrations of ghrelin corrected for BF. However, there is a need for further studies to identify the underlying mechanism and potential implications of such an association.

The mechanisms responsible for the association between infant weight gain and later body composition and appetite regulatory hormones are not clear. In infancy there is an intensive development of adipose tissue (29, 42), and the adipocyte number continues to increase until adolescence (43). Animal studies (44, 45) suggested that perinatal overfeeding may lead to rapid weight gain and later overweight, perhaps by producing fundamental changes in neurologic control of metabolism. However, it may also be a genetic predisposition to overweight, which is expressed early by a rapid weight gain (46) that is not adjusted for by parental BMI, or early weight gain may express the individual’s genetic growth potential (47). Analysis on the basis of the Fels Longitudinal Study showed that genetic factors had strong influence on Δweight-SDS from 0 to 2 y of age (48) and stature in infancy and later life (49, 50).

The strengths of our study were that we assessed BF with DXA scanning and had long term follow-up with examinations before and after puberty. Also, in this study we investigated the effect of weight gain in healthy term AGA infants and examined if the associations shown in studies that also included low BW infants were also present in infants with a normal BW. A limitation of the current study was the small sample size, but despite this, we showed several interesting results.

In conclusion, the consistency of the effect of rapid weight gain in infancy on body composition and concentrations of ghrelin and adiponectin corrected for BF at 17 y of age further support that early weight gain is associated with overweight in later life. However, we did not find that early weight gain programs later IR or leptin. Although this has been shown in preterm infants, such an effect might not exist in term AGA infants.

We gratefully acknowledge Majken Ege for collecting the data and Vivian Anker and Birgitte Hermansen for technical assistance.

The authors’ responsibilities were as follows—AL and KFM: responsible for the study idea; KFM, CM, and HKI: responsible for the study design; HKI: obtained the funding and managed the data collection; JJH: performed the assays on the hormones; AL: conducted the statistical analyses and prepared the first draft of the manuscript; and all authors: participated in interpreting the results and involved in preparing the final draft of the manuscript. None of the authors reported financial or personal conflicts of interest.

FOOTNOTES

2

Supported by the Lundbeck Foundation, the Danish Heart Association, and the Ville Heises Foundation.

REFERENCES

1.

Parsons

TJ

Childhood predictors of adult obesity: a systematic review

.

Int J Obes Relat Metab Disord

1999

;

23

(

suppl 8

):

S1

–

107

.

2.

Curhan

GC

Birth weight and adult hypertension, diabetes mellitus, and obesity in US men

.

Circulation

1996

;

94

:

3246

–

50

.

3.

Oken

E

Fetal origins of obesity

.

Obes Res

2003

;

11

:

496

–

506

.

4.

Gillman

MW

The first months of life: a critical period for development of obesity

.

Am J Clin Nutr

2008

;

87

:

1587

–

9

.

5.

Baird

J

Being big or growing fast: systematic review of size and growth in infancy and later obesity

.

BMJ

2005

;

331

:

929

.

6.

Singhal

A

Low nutrient intake and early growth for later insulin resistance in adolescents born preterm

.

Lancet

2003

;

361

:

1089

–

97

.

7.

Arenz

S

Breast-feeding and childhood obesity–a systematic review

.

Int J Obes Relat Metab Disord

2004

;

28

:

1247

–

56

.

8.

Gillman

MW

Risk of overweight among adolescents who were breastfed as infants

.

JAMA

2001

;

285

:

2461

–

7

.

9.

Agostoni

C

Growth patterns of breast fed and formula fed infants in the first 12 months of life: an Italian study

.

Arch Dis Child

1999

;

81

:

395

–

9

.

10.

Dewey

KG

Growth of breast-fed and formula-fed infants from 0 to 18 months: the DARLING Study

.

Pediatrics

1992

;

89

:

1035

–

41

.

11.

Proietto

J

Animal models of obesity–theories of aetiology

.

Baillieres Clin Endocrinol Metab

1994

;

8

:

509

–

25

.

12.

Rolland-Cachera

MF

Influence of macronutrients on adiposity development: a follow up study of nutrition and growth from 10 months to 8 years of age

.

Int J Obes Relat Metab Disord

1995

;

19

:

573

–

8

.

13.

Lissner

L

Birth weight, adulthood BMI, and subsequent weight gain in relation to leptin levels in Swedish women

.

Obes Res

1999

;

7

:

150

–

4

.

14.

Chanoine

JP

Ghrelin in growth and development

.

Horm Res

2005

;

63

:

129

–

38

.

15.

Bunt

JC

Cross-sectional and prospective relationships of fasting plasma ghrelin concentrations with anthropometric measures in pima Indian children

.

J Clin Endocrinol Metab

2003

;

88

:

3756

–

61

.

16.

Giapros

V

Serum adiponectin and leptin levels and insulin resistance in children born large for gestational age are affected by the degree of overweight

.

Clin Endocrinol (Oxf)

2007

;

66

:

353

–

9

.

17.

Stefan

N

Plasma adiponectin concentrations in children: relationships with obesity and insulinemia

.

J Clin Endocrinol Metab

2002

;

87

:

4652

–

6

.

18.

Singhal

A

Early nutrition and leptin concentrations in later life

.

Am J Clin Nutr

2002

;

75

:

993

–

9

.

19.

Michaelsen

KF

Nutrition and growth during infancy. The Copenhagen Cohort Study

.

Acta Paediatr Suppl

1997

;

420

:

1

–

36

.

20.

Wold Health Organization

.

The WHO Child Growth Standards;

WHO Antro

2005

.

.

21.

Cole

TJ

Establishing a standard definition for child overweight and obesity worldwide: international survey

.

BMJ

2000

;

320

:

1240

–

3

.

22.

Monteiro

PO

Rapid growth in infancy and childhood and obesity in later life–a systematic review

.

Obes Rev

2005

;

6

:

143

–

54

.

23.

Ong

KK

Association between postnatal catch-up growth and obesity in childhood: prospective cohort study

.

BMJ

2000

;

320

:

967

–

71

.

24.

Ong

KK

Size at birth, postnatal growth and risk of obesity

.

Horm Res

2006

;

65

(

suppl 3

):

65

–

9

.

25.

Karaolis-Danckert

N

Rapid growth among term children whose birth weight was appropriate for gestational age has a longer lasting effect on body fat percentage than on body mass index

.

Am J Clin Nutr

2006

;

84

:

1449

–

55

.

26.

Eriksson

M

Associations of birthweight and infant growth with body composition at age 15–the COMPASS study

.

Paediatr Perinat Epidemiol

2008

;

22

:

379

–

88

.

27.

Gardner

DS

Contribution of early weight gain to childhood overweight and metabolic health: a longitudinal study (EarlyBird 36)

.

Pediatrics

2009

;

123

:

e67

–

73

.

28.

Stettler

N

Weight gain in the first week of life and overweight in adulthood: a cohort study of European American subjects fed infant formula

.

Circulation

2005

;

111

:

1897

–

903

.

29.

Botton

J

Postnatal weight and height growth velocities at different ages between birth and 5 y and body composition in adolescent boys and girls

.

Am J Clin Nutr

2008

;

87

:

1760

–

8

.

30.

Leunissen

RW

Timing and tempo of first-year rapid growth in relation to cardiovascular and metabolic risk profile in early adulthood

.

JAMA

2009

;

301

:

2234

–

42

.

31.

Cianfarani

S

Adiponectin levels are reduced in children born small for gestational age and are inversely related to postnatal catch-up growth

.

J Clin Endocrinol Metab

2004

;

89

:

1346

–

51

.

32.

Ong

KK

Birth weight, infant growth and insulin resistance

.

Eur J Endocrinol

2004

;

151

(

suppl 3

):

U131

–

9

.

33.

Finken

MJ

Preterm birth and later insulin resistance: effects of birth weight and postnatal growth in a population based longitudinal study from birth into adult life

.

Diabetologia

2006

;

49

:

478

–

85

.

34.

Ekelund

U

Association of weight gain in infancy and early childhood with metabolic risk in young adults

.

J Clin Endocrinol Metab

2007

;

92

:

98

–

103

.

35.

Iniguez

G

Adiponectin levels in the first two years of life in a prospective cohort: relations with weight gain, leptin levels and insulin sensitivity

.

J Clin Endocrinol Metab

2004

;

89

:

5500

–

3

.

36.

Tamakoshi

K

Low birth weight is associated with reduced adiponectin concentration in adult

.

Ann Epidemiol

2006

;

16

:

669

–

74

.

37.

Kim

CS

Does birthweight predict adolescent adiponectin levels?

Clin Endocrinol (Oxf)

2006

;

64

:

162

–

8

.

38.

Yamamoto

Y

Adiponectin, an adipocyte-derived protein, predicts future insulin resistance: two-year follow-up study in Japanese population

.

J Clin Endocrinol Metab

2004

;

89

:

87

–

90

.

39.

Phillips

DI

Size at birth and plasma leptin concentrations in adult life

.

Int J Obes Relat Metab Disord

1999

;

23

:

1025

–

9

.

40.

Yousheng

Jia

Programmed alterations in hypothalamic neuronal orexigenic responses to ghrelin following gestational nutrient restriction

.

Reprod Sci

2008

;

15

:

702

–

9

.

41.

Yoshinaga

M

Adipokines and the prediction of the accumulation of cardiovascular risk factors or the presence of metabolic syndrome in elementary school children

.

Circ J

2008

;

72

:

1874

–

8

.

42.

Soriguer Escofet

FJ

Adipose tissue fatty acids and size and number of fat cells from birth to 9 years of age–a cross-sectional study in 96 boys

.

Metabolism

1996

;

45

:

1395

–

401

.

43.

Spalding

KL

Dynamics of fat cell turnover in humans

.

Nature

2008

;

453

:

783

–

7

.

44.

Plagemann

A

Perinatal elevation of hypothalamic insulin, acquired malformation of hypothalamic galaninergic neurons, and syndrome x-like alterations in adulthood of neonatally overfed rats

.

Brain Res

1999

;

836

:

146

–

55

.

45.

Plagemann

A

Observations on the orexigenic hypothalamic neuropeptide Y-system in neonatally overfed weanling rats

.

J Neuroendocrinol

1999

;

11

:

541

–

6

.

46.

Stettler

N

Rapid weight gain during infancy and obesity in young adulthood in a cohort of African Americans

.

Am J Clin Nutr

2003

;

77

:

1374

–

8

.

47.

van Dommelen

P

Genetic study of the height and weight process during infancy

.

Twin Res

2004

;

7

:

607

–

16

.

48.

Demerath

EW

Genetic and environmental influences on infant weight and weight change: the Fels Longitudinal Study

.

Am J Hum Biol

2007

;

19

:

692

–

702

.

49.

Towne

B

Genetic analysis of patterns of growth in infant recumbent length

.

Hum Biol

1993

;

65

:

977

–

89

.

50.

Czerwinski

SA

Genetic factors in physical growth and development and their relationship to subsequent health outcomes

.

Am J Hum Biol

2007

;

19

:

684

–

91

.

© 2010 American Society for Nutrition

© 2010 American Society for Nutrition

Topic:

• body mass index procedure
• insulin resistance
• adolescent
• hormones
• body composition
• child
• follow-up
• infant
• leptin
• weight gain
• desire for food
• appetite or desire
• adiponectin
• ghrelin
• body fat
• older adult
• trunk structure

Citations

Views

Altmetric

Email alerts

More on this topic

Related articles in PubMed

Citing articles via

• Latest

• Most Read

• Most Cited

More from Oxford Academic

Which hormones are most responsible growth during infancy and childhood?

What hormone causes growth during infancy?

What hormones are related to growth and development?

What hormone stimulates growth during childhood?