Sex Differences in Infant Body Composition Emerge in the First 5 Months of Life

Abstract

Background:

Sex differences in body composition are appreciated throughout the lifespan with probable contributions from sex steroids: testosterone and estrogen. The purpose of this longitudinal observational study was to determine if sex differences in body composition emerge during the first months of life in healthy infants, corresponding to the age at which male infants are producing endogenous testosterone.

Methods:

Linear growth and body composition parameters using air displacement plethysmography were obtained on 602 healthy infants after birth and again at 5 months of age. Rate of change in body composition parameters were compared between sexes.

Results:

Sex differences in length, total mass, fat free mass (FFM), and percent fat mass (%FM) were present both at birth and 5 months (p<0.001 for all), with males having greater total mass and FFM but lower %FM. Gain in %FM over the first 5 months was significantly lower in males (p=0.0004). This difference was secondary to a gain of 17 grams per week more in FFM in males compared to females.

Conclusion:

Sex differences in body composition are emerging in the first months of life, with lower adiposity accumulation in males. Endogenous testosterone production in males ^~1-4 months of age may account for findings and may have lifelong implications for sex differences in body composition.

Background

Sex differences in body composition are recognized throughout the lifespan, with females having significantly lower lean mass and greater percent fat mass (%FM) than males. Sex steroids (testosterone and estrogen) are assumed to be largely responsible for these differences; however, sex differences are observed in childhood well before puberty occurs. While some studies report that sex differences are present at birth (1-3) others have found no differences between males and females (4, 5). By one year of age, females have greater %FM and this difference seems to persist until puberty at which time it becomes further exaggerated. Although sex differences in body composition are known to exist, the underlying etiology and timing of emergence is surprisingly understudied.

The mini-puberty period of infancy is a biologically plausible time during which sex differences in body composition may emerge. The mini-puberty period refers to temporary activation of the hypothalamic-pituitary-gonadal axis resulting in near-adult concentrations of testosterone in boys from around 1-4 months of age (6). Testosterone is an anabolic steroid that acts to increase protein synthesis and lean mass while decreasing fat mass. Although the existence of the mini-puberty period is well known by pediatric endocrinologists, the short and long-term consequences of this temporary testosterone exposure are just beginning to be unveiled. It was recently shown that there are sex differences in linear growth velocity during the first 6 months of life associated with testosterone concentrations (7). This difference in infancy was calculated to be substantial enough to account for 15% of the final height differences between men and women (8). Animal models have found that blocking the post-natal testosterone surge in male mice results in greater adiposity lifelong (9). The effects sex steroid exposure during the mini-puberty period on body composition in humans have not been studied.

The objective of this secondary data analysis was to determine if sex differences in body composition emerge during the first months of life at a sensitive time period at which sex steroid exposure is known to be substantially different between the sexes. We hypothesize that differences in body composition emerge during the first 5 months of life, with males developing more lean mass and lower percent fat mass in a pattern consistent with testosterone exposure.

Methods

Between 2009 and 2014, the Healthy Start Study enrolled 1410 mother-offspring dyads before 24 weeks gestation from obstetrics clinics at the University of Colorado Hospital. Detailed assessment of the study participants and methods have been previously published (10-13). In brief, sociodemographic data, maternal medical history, pregnancy and birth history, and infant feeding methods were collected. Infant growth and body composition parameters were obtained at birth and again 4-6 months of age. For this analysis, infants had to be born term (≥37 weeks gestation) and have complete data on body composition available for both the neonatal visit (<3 days of age) and the infant visit (~5 months of age, range 3-7 months). The study was approved by the Colorado Multiple Institutional Review Board and all women provided written informed consent for study participation. The Healthy Start Study was registered at clinicaltrials.gov ().

Mothers self-reported their race/ethnicity, tobacco use, and infant feeding status. Maternal race was categorized as Non-Hispanic White, Hispanic, Non-Hispanic Black, and Other which included Asian or Pacific Islander and American Indian or Alaskan Native. The total number of breastfeeding months was calculated as previously described (11). Pre-pregnancy body mass index (BMI), maternal weight gained during pregnancy, gestational age at birth, and infant birthweight were abstracted from medical records.

Infant length was measured supine on a length board to the nearest tenth of a centimeter (cm) by two trained research personnel. Body composition was assessed using air displacement plethysmography (PEA POD, COSMED, Rome, Italy). The PEA POD is a validated instrument that measures total body mass in grams, total body volume, and estimates fat mass (FM), fat-free mass (FFM), and percent fat mass (%FM)(14). Each infant was measured twice, with a third measurement obtained when %FM differed by >2.0%; the average of the two closest readings was used for analysis. The rate of change for each of the following parameters: length, %FM, FM, and FFM, was calculated as the measurement at ~5 months minus the measurement at birth divided by the amount of time between measures in weeks.

Statistical Analysis

Descriptive statistics were used to summarize baseline characteristics of the sample as well as body composition parameters at birth and 5 months of age stratified by sex. Change in body composition parameters between time points were assessed with a paired t-test. Sex differences were evaluated using t-tests for continuous variables and the Cochran-Mantel-Haenszel test for categorical variables.

Separate general linear univariate multivariable models were fit for each outcome with sex of the offspring as the main predictor. A priori covariates included race and the measure of the outcome near birth (i.e., % fat mass near birth was included as covariate when rate of change of % fat mass was the outcome). Although several other covariates are known to affect infant adiposity, including maternal BMI, weight gain during pregnancy, smoking, and infant feeding source, these variables are not associated with infant sex or on the causal pathway between sex and change in body composition so they were not included in the model. We assessed the significance of the covariates using an F test at an alpha level of 0.05. We examined the studentized residuals to ensure we met the model assumptions of normality and homoscedasticity. All model assumptions were met and there were no overly influential observations. The final model for each outcome included race and the measure of the outcome at birth as covariates. The estimates of association, p-values and 95% confidence intervals are presented. Statistical analyses were conducted in SAS 9.4 (SAS Institute, Cary, NC).

Results

Out of the full 1,410 infant cohort, we excluded infants who did not have outcome measures at both time points (N=673), as well as infants born before 37 weeks gestation (N=11), infants who had their initial PEA POD measurement after 3 days of age, or second PEA POD more than 209 days after birth (N=124). Therefore, the analytic cohort included the remaining N=602 participants. Maternal and infant characteristics were similar between the full cohort and the analytic cohort, similar to that previously reported.(11) The mean age of the cohort at the first visit was 1.5 days and second visit was 5 months. Demographic data were similar between males and females (Table 1). Sex differences in length, FM, FFM, and %FM were present at birth for all parameters (p<0.001, Table 2). As expected, length, FM, FFM, and %FM all increased between birth and 5 months in both males and females (p<0.001 for all, Table 1). At 5 months of age, FFM and %FM remained significantly different between males and females, with relative total body adiposity now 8.7% lower in males compared to females ((25.47-23.25)/25.47, p<0.0001). FM was no longer significantly different between sexes at 5 months of age (p=0.12).

Table 1.

Participant Demographics (mean ± SD, n (%))

	Total Sample (n=602)	Females (n=306)	Males (n=296)
Maternal Characteristics
Race
Hispanic	147 (24%)	74 (24%)	73 (25%)
Non-Hispanic white	341 (57%)	173 (57%)	168 (57%)
Non-Hispanic black	83 (14%)	42 (14%)	41 (14%)
Other	31 (5%)	17 (6%)	14 (5%)
Pre-pregnancy BMI (kg/m2)	25.84 ± 6.36	25.68 ± 6.20	26.00 ± 6.52
Gestational weight gain (kg)	13.65 ± 6.38	13.65 ± 6.68	13.65 ± 6.06
Smoking during pregnancy (n)	38 (6%)	18 (6%)	20 (7%)
Infant characteristics
Gestational age (weeks)	39.54 ± 1.15	39.57 ± 1.16	39.50 ± 1.13
Birthweight (grams)	3277 ± 426	3220 ± 410	3330 ± 428
Age at PEA POD #1 (days)	1.12 ± 0.53	1.17 ± 0.56	1.08 ± 0.49
Age at PEA POD #2 (months)	4.94 ± 0.93	4.98 ± 0.92	4.89 ± 0.94
Breastfeeding months	3.48 ± 1.81	3.47 ± 1.80	3.49 ± 1.83

Table 2.

Body composition measures (mean ± SD) at birth and 5 months by infant sex

	Total Sample (n=602)	Females (n=306)	Males (n=296)	p-value
Body composition at birth
Total Body Mass (kg)	3.12 ± 0.41	3.07 ± 0.40	3.17 ± 0.41	0.001
Length (cm)	49.18 ± 2.04	48.75 ± 1.97	49.63 ± 2.02	<0.0001
Fat Free Mass (kg)	2.83 ± 0.33	2.76 ± 0.31	2.91 ± 0.33	<0.0001
Fat Mass (kg)	0.29 ± 0.14	0.31 ± 0.14	0.27 ± 0.13	0.0002
% Fat Mass	9.02 ± 3.72	9.83 ± 3.85	8.18 ± 3.39	<0.0001
Body composition at 5 months
Total Body Mass (kg)	6.74 ± 0.83	6.57 ± 0.82	6.92 ± 0.81	<0.0001
Length (cm)	63.74 ± 2.66	63.09 ± 2.54	64.44 ± 2.62	<0.0001
Fat Free Mass (kg)	5.08 ± 0.57	4.88 ± 0.53	5.29 ± 0.54	<0.0001
Fat Mass (kg)	1.66 ± 0.50	1.70 ± 0.50	1.63 ± 0.50	0.12
% Fat Mass	24.38 ± 5.42	25.47 ± 5.32	23.25 ± 5.30	<0.0001

Sex differences in rate of change per week in body compositon parameters are shown in Table 3. On average, after adjusting for race/ethnicity, FFM increased 17 grams more per week in males compared to females (p<0.0001, 95% CI: 14, 20), yielding a difference of 410 grams of FFM between sexes at 5 months. Rate of FM gain was not significantly different between sexes. For males, %FM increased 0.09 percentage points less per week than females (p=0.0004, 95% CI: 0.04, 0.13), which equates to an 8% difference between sexes in adiposity gain during the first 5 months of life. Length increased 0.05 cm more per week (2.6 cm/yr) in males compared to females (p<0.0001, 95% CI: 0.04, 0.07) during the first 5 months of life.

Table 3.

Sex differences in rate of change week in body composition parameters

Outcome	Beta Coefficient (Males Relative to Females)	Standard Error	P-Value
Rate of change in percent fat mass (%)	−0.09	0.02	0.0004
Rate of change in fat free mass (grams/wk)	16.59	1.65	<0.0001
Rate of change in fat mass (grams/wk)	−1.07	2.00	0.59
Rate of change in length (cm/wk)	0.05	0.01	<0.0001

Discussion

In this large, ethnically diverse cohort of term infants, we have shown that while sex differences in body composition are already present at birth, the magnitude of these differences increases during the first 5 months of life. Gain in adiposity was 8% lower in males compared to females, and this was due to a greater gain in absolute FFM in males vs females. We also confirmed that linear growth velocity is greater in males during this time period. These early differences in growth patterns between males and females are important to consider as we further define the critical role of programming in the early post-natal period on future risk for adult diseases (15).

Sex differences in adiposity are well recognized in adults, with males having less total adipose tissue but disproportionately more visceral adipose tissue (16). This sexual dimorphism in body composition has been demonstrated even in pre-pubertal children, although to a much smaller extent than after puberty (17, 18). Adipose tissue was once thought to be a storage depot for lipids but it is now recognized as an endocrine organ that regulates metabolism (19). Adipose tissue dysfunction is strongly associated with systemic inflammation, insulin resistance, and cardiovascular disease. Furthermore, accumulation of adiposity in the early infancy period has been associated with later childhood, adolescent and adult obesity, supporting the model that programming during critical windows of development and plasticity influence lifelong metabolism (20). Despite this, there has been minimal investigation comparing differences in early growth between sexes, and our understanding of the origins of later sex differences are largely speculative. This study specifically compares longitudinal changes in body composition during this critical time period between males and females.

The mini-puberty period of infancy refers to temporary activation of the hypothalamic-pituitary-gonadal axis and production of sex steroids (testosterone and estrogen) in infancy (6). The time course is more well-defined in males, peaking around 6-8 weeks of age and typically ending by 4-5 months of age. Testosterone is an anabolic hormone that increases lean mass as well as linear growth velocity. Kiviranta et al found sex differences in growth velocity in 18,570 infants from the United Kingdom, with males growing 2-4 cm/yr more than females in the first 6 months of life and no sex differences in growth velocity after 6 months of age (7). Our data corroborate these findings with a difference between sexes of 2.6 cm/yr in the first 5 months of life. These results support the importance of the first few months of life in establishing early sex differences in height. Beyond height, the body composition changes we observed in males in this study are identical to what is expected from testosterone exposure: greater FFM, lower %FM, and higher linear growth velocity compared to females. To confirm these observed changes in body composition parameters were not secondary to the change in length the analyses were repeated with change in length as a covariate in the model and outcomes were unchanged. Although we cannot definitively conclude that these differences are due to testosterone, the body composition changes we observed during the time course of the normal mini-puberty period are consistent with the known effects of testosterone. The implications of the mini-puberty period and it’s coincidental timing with the hypothesized critical programming in the first months of life are just beginning to be investigated.

The major strengths of this study are the large, ethnically diverse cohort of infants with a rigorous assessment of body composition. Limitations of this study include no measures to directly attribute the observed sex differences to testosterone and data limited at this time to the first 6 months of life (though additional follow up is underway). In addition, as the PEA POD assesses total body adiposity we are unable to evaluate sex differences in adiposity depots, such as visceral versus subcutaneous fat, which is recognized to be sex specific and have strong correlates to cardiometabolic disease states in adults (21). Despite these limitations, this study is the first to demonstrate sex differences in body composition are increasing during these early months of life and provide justification for further investigation into the mechanisms and clinical implications of these findings.

In conclusion, our study confirms there are sex differences emerging in the rate of linear growth during early infancy, and also provides novel data supporting emerging sex differences in body composition. Sex differences in body composition are present at birth but significantly widen over the first 5 months of life. FFM increases significantly more in males than females, while FM gain is similar between sexes resulting in lower adiposity gain in males. Although we cannot confirm this is secondary to the testosterone surge occurring in healthy male infants during this time period with our dataset, this pattern of change in body composition is congruent with what would be expected from testosterone exposure. The finding supports further study of the physical sex differences emerging during the mini-puberty period of infancy that may have lifelong implications in health and disease.

List of Abbreviations

FM

fat mass

FFM

fat free mass

%FM

percent fat mass

kg

kilograms

cm

centimeters

yr

year

wk

week

CI

confidence interval