Summary
Objective
Little is known about the relationships of dietary factors, physical activity, and sedentary behavior to dehydroepiandrosterone sulfate (DHEAS) and insulin-like growth factor-1 (IGF-1) concentrations among prepubertal children. Therefore, we studied the associations of these lifestyle factors with serum DHEAS and IGF-1 in children.
Design and subjects
Cross-sectional analysis of a population sample of 431 prepubertal children aged 6-9 years.
Measurements
Assessment of dietary factors by food records and physical activity and sedentary behavior by a combined heart rate and movement monitor and a questionnaire. Measurement of serum DHEAS and IGF-1.
Results
Consumption of low-fiber grain products (standardized regression coefficient β=0.118, P=0.017) and intake of vegetable protein (β=0.100, P=0.045) were positively and consumption of sugar-sweetened beverages (β=-0.117, P=0.018) was inversely associated with DHEAS after adjustment for sex, age, and body fat percentage. Energy intake (β=0.160, P=0.001) was positively associated with IGF-1 adjusting for sex, age, and body fat percentage. Vigorous physical activity was inversely associated with DHEAS after adjustment for sex and age (β=-0.120, P=0.027), and total (β=-0.137, P=0.007), moderate (β=-0.130, P=0.012), vigorous (β=-0.136, P=0.011), and moderate to vigorous physical activity (β=-0.160, P=0.003) were inversely and total sedentary behavior (β=0.151, P=0.003) was positively associated with IGF-1 adjusting for sex and age. None of physical activity measures was associated with DHEAS or IGF-1 after additional adjustment for body fat percentage.
Conclusions
Lifestyle factors have weak and moderate associations with biochemical markers of adrenarche in prepubertal children. These associations indicate body fat independent and dependent influences of diet and physical activity, respectively.
Keywords: DHEAS, IGF-1, Body fat percentage, Lean body mass, Lifestyle factors
Introduction
Adrenarche is a gradually developing process of the adrenal cortex leading to increased production of androgen precursors, mainly dehydroepiandrosterone (DHEA) its sulfate (DHEAS), and androstenedione (1). These weak androgens are converted peripherally to more potent androgens, testosterone and dihydrotestosterone, leading to clinical signs of adrenarche, including adult type body odor, oily hair, acne or comedones, pubic or axillary hair, and increased statural growth (2).The timing of adrenarche varies widely between children and appears to be influenced by prenatal and postnatal factors (3). In turn, the biochemical markers of adrenarche have been associated with earlier age at menarche (4).
The weight-adjusted heritability of adrenal androgen secretion is high, but also environmental factors play a role (5). The association of adiposity with adrenarche has been shown in several studies (6–8). Body fat percentage was found to correlate positively with serum DHEAS concentration (9), and body mass index (BMI) change was observed to correlate positively with urinary excretion of DHEAS in children (10). Furthermore, we showed recently that adiposity increased the likelihood of clinical signs of adrenarche in a population sample of children aged 6-9 years (11).
Prepubertal children with premature adrenarche have been found to have elevated serum IGF-1 levels (12). IGFs and their receptors are expressed in human adrenal cortex (13, 14), and IGF-1 can increase adrenal androgen production (15). However, there is no evidence that IGF-1 would directly increase adrenocortical androgen production during adrenarche (14).
Increased dietary intake of vegetable protein and fiber have been associated with decreased gonadal estradiol production in pubertal girls independent of body size (16). However, there are only few studies on the associations of dietary factors with adrenal androgen metabolism in children. In one previous study, a higher intake of dietary animal protein was associated with a higher urinary excretion of C19 steroid metabolites that was used as a marker of adrenal androgen secretion (7). Interestingly, some animal studies suggest that the administration of DHEA decreases the intake of food containing large amounts of fat in lean and obese rats (17). This effect was mediated by hypothalamic neurotransmitters (18). Moreover, a higher intake of protein(19, 20), animal protein (19, 21), and energy (20), a lower intake of fat, monounsaturated fat, and polyunsaturated fat (19), and a higher consumption of milk (21, 22) have been related to a higher serum IGF-1 concentration in healthy children.
The results of some earlier studies suggest that increased physical activity is associated with decreased serum testosterone levels in boys independent of BMI (23). We are not aware of any published reports on the association of physical activity or sedentary behavior with serum or urinary adrenal androgen levels in children. However, there is some evidence that a single bout of exercise increases serum IGF-1 concentration (24) but that exercise training during negative energy balance or heavy training decreases serum IGF-1 concentrations in children and adults (24–26). There is no evidence on the association between sedentary behavior and serum IGF-1 concentration in children.
Little is known about the associations of dietary factors, physical activity, and sedentary behavior with serum DHEAS and IGF-1 concentrations among prepubertal children independent of body composition. We therefore examined these relationships in a population sample of children aged 6-9 years.
Subjects and Methods
Study population
These analyses are based on data from the baseline examinations of the Physical Activity and Nutrition in Children (PANIC) Study which is an ongoing controlled physical activity and dietary intervention study in a population sample of primary school children from the city of Kuopio in Finland (27).
Altogether 736 children aged 6-9 years from 16 schools of the city of Kuopio were invited to participate in the study. Of them, 512 (70 %) participated in the baseline examinations in 2007-2009. The participants did not differ in age, sex distribution, or BMI standard deviation score (BMI-SDS) from all children who started the first grade in the primary schools of Kuopio in 2007–2009 based on comprehensive data obtained from school health examinations. The exclusion criteria for the analyses were central puberty and long-term medication that could have an effect on growth or adrenal function. Data on food consumption were available for 395 children (190 girls, 205 boys) and data on physical activity and sedentary behavior for 431 children (207 girls, 224 boys). The study was conducted according to the guidelines laid down in the Declaration of Helsinki. The study protocol was approved by the Research Ethics Committee of the Hospital District of Northern Savo. All children participating in the study and their parents gave their informed written consent.
Assessment of pubertal signs, DHEAS, and IGF-1
Central gonadotropin-dependent puberty was defined by clinical examination as breast development at Tanner stage ≥2 for girls and testicular volume ≥4 mL assessed using an orchidometer for boys.
All serum samples were taken after an overnight fast and kept deep frozen until they were analyzed. Serum DHEAS concentrations were determined using an enzyme-linked immunosorbent assay ELISA kit (Alpha Diagnostic International, San Antonio, TX). The intra-assay coefficient of variation (CV) of the DHEAS method was 7.5-11.5% and the inter-assay CV was 7.0-11.0%. Serum IGF-1 concentrations were determined using an ELISA kit (Mediagnost, Reutlingen, Germany). The intra-assay CV of the IGF-1 method was 5.1-6.6% and the inter-assay CV was 7.7-9.2%. We defined biochemical adrenarche as serum DHEAS concentration ≥ 40 μg/dL (1.08 μmol/L) (28) and premature adrenarche as biochemical adrenarche with any clinical sign of adrenarche (adult type body odor, oily hair, acne or comedones, or pubic or axillary hair) before eight years in girls and nine years in boys (2).
Assessment of body size and composition
Body weight was measured twice after an overnight fast, bladder emptied, and standing in light underwear by InBody ®720 bioelectrical impedance device (Biospace Co. Ltd., Seoul, Korea) to accuracy of 0.1 kilograms. The mean of the two values was used in the analyses. Body height was measured three times in the Frankfurt plane without shoes using a wall-mounted stadiometer to accuracy of 0.1 centimeter. The mean of the two nearest values was used in the analyses. BMI was calculated as body weight (kg) divided by body height (m) squared. Height standard deviation score (Height- SDS), BMI-SDS, and the prevalence of overweight were calculated according to the Finnish growth references (29). Body fat percentage and lean body mass (kg) were measured bladder emptied and lying in light clothing by the Lunar dual-energy X-ray absorptiometry (DXA) device (Lunar Prodigy Advance; GE Medical Systems, Madison, WI). Information on gestational age at birth, birth weight, and birth length were obtained from Kuopio University Hospital records.
Assessment of dietary factors
Eating frequency, food consumption, and total energy and nutrient intakes were assessed by food records of either consecutive four days, including at least one weekend day (91.6%), or consecutive three days, including at least one weekend day (8.4%). A clinical nutritionist instructed the parents to record all food and drinks consumed by their child at home, at school, in afternoon care, and elsewhere outside home using household or other measures, such as tablespoons, deciliters, and centimeters, in person at the first study visit (27). The parents were instructed to report the recipes of mixed dishes and the brands and the contents of food products. A clinical nutritionist reviewed the food records with the parents at the second study visit and completed the records using a picture booklet of portion sizes. Moreover, a clinical nutritionist asked the catering company about the details of food and drinks, such as menus, cooking fat, and spread on bread, served at schools and in afternoon care. All prepared foods and mixed dishes were disaggregated into ingredients according to the recipes used. Meals were defined according to the recorded time and type of food and were classified as breakfast, lunch, and dinner, and all eating and drinking occasions between the meals were classified as snacks. We analysed food consumption and nutrient intake using The Micro Nutrica dietary analysis software, version 2.5 (The Social Insurance Institution of Finland), that uses Finnish and international data on the nutrient compositions of foods (30). A clinical nutritionist also updated the software by adding new food items and products with their precise nutrient content received from the producers.
Assessment of physical activity and sedentary behavior
Physical activity and sedentary behavior were assessed by a combined heart rate and movement monitor (Actiheart, CamNtech, Cambridge, UK) for four consecutive days without interruption (31). The monitor was attached on the chest with two standard electrocardiogram electrodes (Bio Protech Inc, Korea). The monitor was set to record in 60-second epochs. Upon retrieving the monitoring device, heart rate data were first cleaned, then individually calibrated with parameters from the cycle test and combined with trunk acceleration using branched equation modeling to produce intensity time-series. Whilst minimizing diurnal bias caused by any potential non-wear episodes, physical activity energy expenditure was calculated by time-integration of the intensity time-series, and the time distribution of activity intensity was generated by using standard metabolic equivalents (METs). For these analyses, the equivalent of 3.5ml O2/min/kg (71 J/min/kg) was used to define 1 MET, and data were summarized as sedentary behavior (≤1.5 METs), light physical activity (>1.5-4 METs), moderate physical activity (>4-7 METs), and vigorous physical activity (>7 METs). Sedentary behavior was calculated by subtracting sleep from total time of ≤1.5METs. Physical activity records were included in the analysis if they contained ≥48h wear data.
Physical activity and sedentary behavior were also assessed by the PANIC Physical Activity Questionnaire filled out by the parents (32). Total daily amounts of physical activity and sedentary behavior were calculated in minutes per day. Types of physical activity included unsupervised physical activity, organized sports, organized exercise other than sports, commuting to and from school, and physical activity during recess. All children in the first grade had 90 minutes of physical education per week at school which was included in total physical activity. Types of sedentary behavior included screen-based sedentary behavior (watching television and videos, using the computer and playing video games, using a mobile phone and playing mobile games), sedentary behavior related to academic skills (reading, writing), sedentary behavior related to music (listening to music, playing a musical instrument), sedentary behavior related to arts, crafts, and games (drawing, doing arts and crafts, playing board and card games), and sitting and lying for a rest.
Statistical methods
The SPSS statistical analysis software, Version 21.0 (IBM Corp., Armonk, NY), was used for statistical analyses. We analyzed the correlations between body fat percentage, lean body mass, and serum DHEAS and IGF-1 concentrations in all children and separately in girls and boys using the non-parametric Spearman`s rank order correlation test. For parametric tests, we natural logarithm transformed serum DHEAS and IGF-1 concentrations to normalize their skewed distributions. We analyzed the associations of dietary factors, physical activity, and sedentary behavior with serum DHEAS and IGF-1 concentrations by linear regression analysis adjusted for sex and age (Model 1) and additionally for body fat percentage (Model 2) or for lean body mass (Model 3). We performed logistic regression analyses adjusted for sex, age, and body fat percentage to explore whether dietary factors, physical activity, and sedentary behavior were associated with the risk of having serum DHEAS concentration ≥ 40 μg/dL (1.08 μmol/L) that has been suggested as definition for biochemical adrenarche (28). Associations with P <0.05 were considered statistically significant.
Results
The characteristics of children are shown in Table 1. The median age of the children was 7.6 years. Almost all children (95%) were born full term and appropriate for gestational age and their median birth weight was 3590 g. Altogether 56 children (13%) were overweight or obese. Biochemical adrenarche indicated by serum DHEAS concentration ≥ 40 μg/dL (1.08 μmol/L) was present in 65 children (15%, 29 girls and 36 boys) and 16 children (4%) had premature adrenarche.
Table 1. Characteristics of children.
| Median | Interquartile range | Range | |
|---|---|---|---|
| Age, years | 7.6 | 7.4-7.9 | 6.6-8.9 |
| Body height, cm | 129.0 | 125.2-132.0 | 110.7-144.7 |
| Body height-SDSa | 0.1 | -0.6-0.8 | -2.8-3.1 |
| Body weight, kg | 25.9 | 23.4-29.0 | 15.2-51.4 |
| BMI-SDSa | -0.2 | -1.0-0.5 | -3.5-2.4 |
| Body fat percentage | 18.7 | 13.3-23.8 | 5.4-44.8 |
| Lean body mass, kg | 20.4 | 18.9-22.2 | 13.3-28.1 |
| Gestational age at birth, wk | 40 | 39-40 | 32-42 |
| Birth length, cm | 50.0 | 49.0-51.0 | 40.5-56.0 |
| Birth weight, g | 3590 | 3203-3880 | 1595-5090 |
| Serum DHEAS, µmol/L | 0.57 | 0.32-0.84 | 0.01-7.84 |
| Serum IGF-1, nmol/L | 22.1 | 17.8-26.8 | 7.76-52.1 |
| Eating frequency | |||
| Meals | 2.8 | 2.5-3.0 | 1.8-3.0 |
| Snacks | 2.5 | 2.0-3.3 | 0.3-8.0 |
| Food consumption | |||
| Red meat, g/MJ | 11.1 | 7.8-14.7 | 0.4-39.7 |
| Poultry, g/MJ | 1.6 | 0.0-3.9 | 0.0-21.6 |
| Fish, g/MJ | 0.8 | 0.0-3.6 | 0.0-16.2 |
| Low-fiber grain products (fiber < 5 %), g/MJ | 15.5 | 11.8-20.4 | 0.4-66.2 |
| High-fiber grain products (fiber ≥ 5 %), g/MJ | 8.5 | 5.1-12.6 | 0.0-42.5 |
| Low-fat milk and sour milk products (fat < 1 %), g/MJ | 59.1 | 14.6-89.9 | 0.0-192.3 |
| High-fat milk and sour milk products (fat ≥ 1 %), g/MJ | 28.3 | 15.4-55.7 | 0.0-169.0 |
| Cheese, g/MJ | 1.7 | 0.6-3.1 | 0.0-11.7 |
| Butter, g/MJ | 0.4 | 0.1-1.4 | 0.0-6.3 |
| Vegetable oil and oil-based margarine, g/MJ | 1.9 | 1.1-3.0 | 0.0-8.8 |
| Vegetables (excluding potato), g/MJ | 13.6 | 8.5-19.1 | 0.9-61.4 |
| Potatoes, g/MJ | 10.1 | 6.3-14.5 | 0.0-43.3 |
| Berries and fruit, g/MJ | 13.5 | 6.6-21.4 | 0.0-73.0 |
| Sugar-sweetened beverages, g/MJ | 15.6 | 6.5-30.1 | 0.0-103.3 |
| Sweets and chocolate, g/MJ | 3.7 | 1.7-6.1 | 0.0-19.6 |
| Nutrient intake | |||
| Energy, MJ/day | 6.8 | 6.0-7.8 | 3.0-10.9 |
| Protein, E% | 16.7 | 15.0-18.4 | 9.5-25.4 |
| Vegetable protein, E% | 4.2 | 3.7-4.8 | 2.3-11.2 |
| Animal protein, E% | 12.3 | 10.5-13.8 | 4.1-21.7 |
| Fat, E% | 29.9 | 26.4-33.6 | 16.1-44.7 |
| Saturated fat, E% | 12.1 | 10.2-14.2 | 5.0-19.4 |
| Monounsaturated fat, E% | 9.8 | 8.7-11.3 | 5.0-18.3 |
| Polyunsaturated fat, E% | 4.7 | 4.0-5.6 | 1.9-10.9 |
| Carbohydrates, E% | 51.5 | 48.3-55.2 | 36.3-66.9 |
| Sucrose, E% | 12.5 | 10.3-14.8 | 0.6-24.5 |
| Fiber, g/MJ | 2.0 | 1.7-2.5 | 0.7-4.1 |
| Physical activity assessed objectively | |||
| Total physical activity, min/day | 660.8 | 561.3-719.9 | 169.3-905.4 |
| Light physical activity, min/day | 527.8 | 442.6-584.7 | 151.0-766.7 |
| Moderate physical activity, min/day | 81.4 | 57.3-114.6 | 9.9-300.9 |
| Vigorous physical activity, min/day | 17.2 | 6.4-35.8 | 0.0-132.1 |
| Moderate to vigorous physical activity, min/day | 103.4 | 71.4-152.1 | 9.9-326.3 |
| Physical activity assessed by questionnaire | |||
| Total physical activity, min/day | 107.9 | 77.1-140.7 | 31.4-247.1 |
| Unsupervised physical activity, min/day | 42.9 | 25.7-68.6 | 0.0-107.1 |
| Organized sports, min/day | 0.0 | 0.0-12.9 | 0.0-51.4 |
| Organized exercise other than sports, min/day | 8.6 | 0.0-17.1 | 0.0-124.3 |
| Commuting to and from school, min/day | 20.0 | 10.0-40.0 | 0.0-180.0 |
| Physical activity during recess, min/day | 20.0 | 20.0-30.0 | 10.0-30.0 |
| Sedentary behavior assessed objectively | |||
| Awake-time sedentary behavior, min/day | 660.8 | 562.9-719.7 | 169.3-869.8 |
| Sedentary behavior assessed by questionnaire | |||
| Total sedentary behavior, min/day | 196.4 | 143.6-263.6 | 19.3-666.4 |
| Screen-based sedentary behavior, min/day | 98.6 | 66.4-128.6 | 0.0-314.3 |
| Sedentary behavior related to academic skills, min/day | 25.7 | 5.7-45.7 | 0.0-180 |
| Sedentary behavior related to music, min/day | 0.0 | 0.0-25.7 | 0.0-205.7 |
| Sedentary behavior related to arts, crafts, and games, min/day | 45.7 | 0.0-77.1 | 0.0-295.7 |
| Sitting and lying for a rest, min/day | 0.0 | 0.0-5.7 | 0.0-162.9 |
DHEAS = dehydroepiandrosterone sulfate, IGF-1 = insulin-like growth factor-1, E% = percentage of total energy intake, SDS = standard deviation score
Height and BMI standard deviation scores based on Finnish reference (29).
Number of children (n) varies from 376 to 431 in different variables:
n= 420: Body fat percentage, Lean body mass
n= 428: Gestational age at birth, Birth weight
n= 421: Birth length
n= 395: Eating frequency, Food consumption, Nutrient intake
n= 430: Physical activity during recess
n= 376: Light physical activity, Moderate physical activity, Vigorous physical activity; objectively measured
n= 374: Awake-time sedentary behavior; objectively measured
Correlations between serum DHEAS, serum IGF-1, body fat percentage, and lean body mass
Body fat percentage correlated positively with serum DHEAS concentration in all children (Table 2), girls (Spearman’s correlation coefficient rho=0.157, P=0.025), and boys (rho=0.180, P=0.008). Lean body mass also correlated positively with serum DHEAS concentration in all children (Table 2), girls (rho=0.264, P<0.001), and boys (rho=0.148, P=0.029). Serum IGF-1 concentration correlated positively with serum DHEAS concentration in all children (Table 2) and girls (rho=0.238, P=0.001) but not boys (rho=0.106, P=0.114). Body fat percentage correlated positively with serum IGF-1 concentration in all children (Table 2), girls (rho=0.210, P=0.003) and boys (rho=0.406, P<0.001), and again similar directions of correlations were observed for lean body mass and serum IGF-1 concentration in all children (Table 2), girls (rho=0.355, P<0.001), and boys (rho=0.269, P<0.001).
Table 2. Correlations between serum DHEAS, serum IGF-1, body fat percentage, and lean body mass.
| Serum IGF-1, nmol/L | Body fat percentage | Lean body mass, kg | |
|---|---|---|---|
| Serum DHEAS, µmol/L | 0.160 (P=0.001) | 0.155 (P=0.001) | 0.197 (P<0.001) |
| Serum IGF-1, nmol/L | 0.363 (P<0.001) | 0.172 (P<0.001) | |
| Body fat percentage | 0.103 (P=0.034) |
Data are correlation coefficients (P-values) from Spearman´s rank order correlation test.
DHEAS = dehydroepiandrosterone sulfate, IGF-1 = insulin-like growth factor-1
Associations of dietary factors with serum DHEAS and IGF-1 concentrations
A higher consumption of low-fiber grain products was associated with a higher serum DHEAS concentration after adjustment for sex and age and after additional adjustment for body fat percentage or lean body mass (Table 3). A lower consumption of sugar-sweetened beverages was related to a higher serum DHEAS concentration after adjustment for sex, age, and body fat percentage or for sex, age, and lean body mass (Table 3). A higher intake of vegetable protein was associated with a higher serum DHEAS concentration after adjustment for sex, age, and body fat percentage (Table 3).
Table 3. Associations of dietary factors with serum DHEAS and IGF-1 concentrations.
| DHEAS, µmol/L | IGF-1, nmol/L | |||||
|---|---|---|---|---|---|---|
| Model 1 | Model 2 | Model 3 | Model 1 | Model 2 | Model 3 | |
| Eating frequency | ||||||
| Mealsa | 0.026 | 0.054 | 0.026 | -0.053 | 0.004 | -0.030 |
| Snacksb | -0.046 | -0.048 | -0.045 | -0.027 | -0.045 | -0.041 |
| Food consumption | ||||||
| Red meat, g/MJ | -0.013 | -0.032 | -0.016 | 0.019 | -0.008 | 0.013 |
| Poultry, g/MJ | -0.076 | -0.084 | -0.071 | 0.036 | 0.020 | 0.044 |
| Fish, g/MJ | 0.050 | 0.054 | 0.051 | 0.028 | 0.033 | 0.028 |
| Low-fiber grain products (fiber <5 %), g/MJ | 0.100* | 0.118* | 0.107* | -0.022 | 0.000 | -0.012 |
| High-fiber grain products (fiber ≥5 %), g/MJ | 0.046 | 0.037 | 0.036 | -0.053 | -0.074 | -0.083 |
| Low-fat milk and sour milk products (fat <1 %), g/MJ | 0.082 | 0.066 | 0.057 | 0.008 | -0.016 | -0.048 |
| High-fat milk and sour milk products (fat ≥1 %), g/MJ | -0.083 | -0.063 | -0.048 | 0.009 | 0.030 | 0.077 |
| Cheese, g/MJ | 0.033 | 0.031 | 0.026 | 0.055 | 0.024 | 0.003 |
| Butter, g/MJ | -0.057 | -0.063 | -0.061 | -0.004 | 0.008 | 0.018 |
| Vegetable oil and oil-based margarine, g/MJ | 0.030 | 0.035 | 0.038 | -0.049 | -0.055 | -0.049 |
| Vegetables (excluding potato), g/MJ | 0.018 | 0.011 | 0.004 | 0.071 | 0.064 | 0.048 |
| Potatoes, g/MJ | -0.038 | -0.043 | -0.034 | -0.006 | -0.019 | -0.006 |
| Berries and fruit, g/MJ | -0.010 | -0.011 | -0.022 | 0.010 | 0.019 | 0.002 |
| Sugar-sweetened beverages, g/MJ | -0.096 | -0.117* | -0.105* | -0.020 | -0.034 | -0.014 |
| Sweets and chocolate, g/MJ | -0.043 | -0.034 | -0.049 | -0.013 | 0.001 | -0.024 |
| Nutrient intake | ||||||
| Energy, MJ/day | 0.023 | 0.014 | -0.023 | 0.149** | 0.160** | 0.090 |
| Protein, E% | 0.035 | 0.022 | 0.031 | 0.091 | 0.059 | 0.066 |
| Vegetable protein, E% | 0.084 | 0.100* | 0.084 | -0.022 | -0.013 | -0.044 |
| Animal protein, E% | 0.000 | -0.020 | -0.004 | 0.090 | 0.057 | 0.075 |
| Fat, E% | -0.053 | -0.055 | -0.042 | -0.031 | -0.034 | -0.005 |
| Saturated fat, E% | -0.078 | -0.078 | -0.065 | -0.020 | -0.019 | 0.011 |
| Monounsaturated fat, E% | -0.051 | -0.051 | -0.033 | -0.044 | -0.055 | -0.018 |
| Polyunsaturated fat, E% | 0.066 | 0.069 | 0.069 | -0.029 | -0.029 | -0.027 |
| Carbohydrates, E% | 0.036 | 0.045 | 0.028 | -0.014 | 0.004 | -0.027 |
| Sucrose, E% | -0.041 | -0.045 | -0.044 | -0.019 | -0.017 | -0.016 |
| Fiber, g/MJ | 0.087 | 0.092 | 0.082 | -0.012 | -0.013 | -0.033 |
Data are standardized regression coefficients from linear regression models adjusted for sex and age (Model 1) and additionally for body fat percentage (Model 2) or lean body mass (Model 3).
DHEAS = dehydroepiandrosterone sulfate, IGF-1 = insulin-like growth factor-1, E% = percentage of total energy intake
*P < 0.05, **P < 0.01. Associations with P < 0.05 are bolded.
Meals coded as 0 = less than three main meals daily, 1= three main meals daily
Snacks coded as 0 = ≤2 snacks/day, 1 = >2 snacks/day
A higher energy intake was related to a higher serum IGF-1 concentration after adjustment for sex and age and after further adjustment for body fat percentage (Table 3).
Associations of physical activity and sedentary behavior with serum DHEAS and IGF-1 concentrations
A lower level of objectively assessed vigorous physical activity was associated with a higher DHEAS after adjustment for sex and age and after additional adjustment for lean body mass (Table 4). Sedentary behavior was not associated with DHEAS (Table 4).
Table 4. Associations of physical activity and sedentary behavior with serum DHEAS and IGF-1 concentrations.
| DHEAS, µmol/L | IGF-1, nmol/L | |||||
|---|---|---|---|---|---|---|
| Model 1 | Model 2 | Model 3 | Model 1 | Model 2 | Model 3 | |
| Physical activity assessed objectively | ||||||
| Total physical activity, min/day | -0.021 | 0.046 | 0.001 | -0.137** | -0.035 | -0.098* |
| Light physical activity, min/day | 0.032 | 0.071 | 0.050 | -0.077 | -0.010 | -0.040 |
| Moderate physical activity, min/day | -0.071 | -0.011 | -0.050 | -0.130* | -0.049 | -0.112* |
| Vigorous physical activity, min/day | -0.120* | -0.075 | -0.121* | -0.136* | -0.040 | -0.129* |
| Moderate to vigorous physical activity, min/day | -0.103 | -0.038 | -0.086 | -0.160** | -0.060 | -0.142** |
| Physical activity assessed by questionnaire | ||||||
| Total physical activity, min/day | -0.026 | 0.002 | -0.048 | -0.059 | -0.006 | -0.091 |
| Unsupervised physical activity, min/day | 0.002 | 0.019 | -0.009 | -0.086 | -0.038 | -0.083 |
| Organized sports, min/day | -0.059 | -0.043 | -0.092 | -0.005 | 0.028 | -0.055 |
| Organized exercise other than sports, min/day | -0.051 | -0.034 | -0.018 | 0.016 | 0.044 | 0.018 |
| Commuting to and from school, min/day | 0.009 | 0.023 | -0.001 | 0.019 | 0.023 | -0.020 |
| Physical activity during recess, min/day | -0.093 | -0.073 | -0.092 | -0.082 | -0.051 | -0.081 |
| Sedentary behavior assessed objectively | ||||||
| Awake-time sedentary behavior, min/day | 0.059 | -0.004 | 0.042 | 0.151** | 0.041 | 0.113* |
| Sedentary behavior assessed by questionnaire | ||||||
| Total sedentary behavior, min/day | 0.018 | 0.015 | 0.026 | 0.103* | 0.076 | 0.093* |
| Screen-based sedentary behavior, min/day | 0.012 | 0.010 | 0.010 | 0.060 | 0.039 | 0.034 |
| Sedentary behavior related to academic skills, min/day | 0.041 | 0.034 | 0.046 | 0.022 | 0.007 | 0.027 |
| Sedentary behavior related to music, min/day | -0.049 | -0.037 | -0.044 | 0.041 | 0.048 | 0.037 |
| Sedentary behavior related to arts, crafts, and games, min/day | 0.026 | 0.012 | 0.027 | 0.088 | 0.070 | 0.094* |
| Sitting and lying for a rest, min/day | 0.002 | 0.014 | 0.030 | 0.071 | 0.050 | 0.075 |
Data are standardized regression coefficients from linear regression models adjusted for sex and age (Model 1) and additionally for body fat percentage (Model 2) or lean body mass (Model 3).
DHEAS = dehydroepiandrosterone sulfate, IGF-1 = insulin-like growth factor-1.* P < 0.05, **P < 0.01. Associations with P < 0.05 are bolded.
A lower level of objectively assessed total physical activity was associated with a higher serum IGF-1 concentration after adjustment for sex and age and after additional adjustment for lean body mass (Table 4). Lower levels of objectively assessed moderate, vigorous, and moderate to vigorous physical activity were associated with a higher serum IGF-1 concentration after adjustment for sex and age and after additional adjustment for lean body mass (Table 4). Higher levels of awake-time sedentary behavior assessed objectively or by a questionnaire were associated with a higher IGF-1 after adjustment for sex and age and after additional adjustment for lean body mass (Table 4). A higher level of sedentary behavior related to arts, crafts, and games was associated with a higher serum IGF-1 concentration after adjustment for sex, age, and lean body mass (Table 4).
Associations of dietary factors, physical activity, and sedentary behavior with the risk of having serum DHEAS concentration ≥ 40 μg/dL (1.08 μmol/L)
Eating frequency, food consumption, nutrient intake, physical activity, or sedentary behavior was not associated with the risk of having serum DHEAS concentration ≥ 40 μg/dL (data not shown).
Discussion
Our study in a population sample of prepubertal children showed that a higher consumption of low-fiber grain products, a lower consumption of sugar- sweetened beverages, a higher intake of vegetable protein, and a lower level of vigorous physical activity were related to a higher serum DHEAS concentration after controlling for body fat percentage. A higher energy intake, lower levels of total, moderate, vigorous, and moderate-to-vigorous physical activity, and a higher level of total sedentary behavior were associated with a higher serum IGF-1 concentration.
We found positive correlations of body fat percentage and lean body mass with serum DHEAS concentration that is in line with the results of a previous report (9). Our observation is also consistent with the notion that the timing of adrenarche is more advanced in children who experience faster postnatal growth (3). Moreover, we found that serum DHEAS concentration was positively correlated with serum IGF-1 concentration in girls but not in boys, which is also consistent with the results of earlier studies (33, 34). We showed recently that prepubertal girls are more likely to have clinical signs of androgen action than prepubertal boys (11). Courant and coworkers (35) showed that prepubertal girls have higher circulating levels of androgen metabolites than boys. Aromatizable androgens converted to estrogens are capable to stimulate the growth hormone - IGF-1 axis (36). Thus, it is possible that DHEAS through its conversion products stimulates growth hormone and IGF-1 production more in girls than in boys. These observations may explain the positive correlation between serum DHEAS and IGF-1 concentration in girls but not in boys.
To our knowledge, there are no published reports on the relationships of dietary factors with serum DHEAS concentration. Such associations could arise by at least two different mechanisms: dietary factors may have an effect on the hypothalamo-pituitary-adrenal axis or adrenal androgens may affect food intake through hypothalamic neurotransmitters, which mechanism has been shown in animal models (17, 18). There is only one report on the association of nutrient intake with adrenal androgen secretion among prepubertal children (7). In this study, a higher intake of dietary animal protein was related to a higher urinary excretion of C19 steroid metabolites that were used as a marker of adrenal androgen secretion (7). We found moderate associations of a higher consumption of low-fiber grain products, a lower consumption of sugar sweetened beverages, and a higher intake of vegetable protein with a higher serum DHEAS concentration. However, we did not observe any association of animal protein intake with serum DHEAS concentration in our population sample of children. This inconsistency may be explained by the different methods used in the evaluation of adrenal androgen secretion or food cultures in the study populations, and as the findings in these two are not consistent with each other, it is possible that apart from affecting body fat content, diet has no major effects on adrenocortical function. To study this further, a controlled study on modulated diet should be conducted. Further, if DHEA(S) affects food intake also in humans, a trial on the effect of DHEA substitution on food intake would be interesting.
The associations of dietary factors with serum IGF-1 concentration is of interest as IGF-1 may mediate some of the effects of diet on growth and health. Rogers and coworkers (19) observed that a higher intake of energy, total protein and animal protein and a lower intake of total fat, monounsaturated fat, and polyunsaturated fat but not measures of food consumption were associated with higher serum IGF-1 concentration in children aged seven to eight years. However, a higher consumption of fruit was related to a higher serum IGF-1 concentration among boys. Similarly, in some other studies a higher intake of energy (20), total protein (20), and animal protein (21) and, a higher consumption of milk (21, 22) have been related to a higher serum IGF-1 concentration in children. We found that a higher energy intake but not eating frequency, energy-adjusted food consumption, or nutrient intake was associated with higher serum IGF-1 levels in children. However, it is difficult to compare the results of these studies with different age ranges and numbers of participants, because the age of children affects dietary intake and serum IGF-1 levels and a small number of children may have lead in false positive findings. It is also possible that some of the inconsistent results in these studies are explained by differences in food cultures, such as sources of animal and vegetable protein, in the study populations.
Acute physical exercise seems to increase circulating DHEA and DHEAS, whereas the effects of exercise training on these have been contradictory in adults (37). However, there are no previous reports on the association of physical activity or sedentary behavior with serum DHEAS concentration in children. We observed that a lower level of vigorous physical activity but not less intensive physical activity or sedentary behavior was associated with a higher serum DHEAS concentration. However, the relationship between vigorous physical activity and DHEAS levels weakened after controlling for body fat percentage but not lean body mass. The explanation for this finding is that body fat percentage is associated not only with DHEAS concentration but also vigorous physical activity (38).
A single bout of exercise has been found to increase serum IGF-1 concentration (24). The reason for this may be a release of IGF-1 from skeletal muscle or from IGF binding proteins. On the contrary, exercise training during negative energy balance or heavy training may decrease serum IGF-1 levels in children (24–26) by decreasing liver IGF-1 production (24). Exercise training may also increase the levels of inflammatory cytokines which can inhibit IGF-1 production (26). However, there are no published reports on the association of physical activity or sedentary behavior with serum IGF-1 concentration in population samples of children. We found that a higher level of moderate-to-vigorous physical activity and a lower level of sedentary behavior were associated with a lower serum IGF-1 concentration in children. These associations were largely explained by body fat percentage that has previously been positively correlated with serum IGF-1 concentration in children (9). However, controlling for lean body mass had less effect on these relationships that may be explained by its weaker association with IGF-1 level in our study and an earlier study (9).
The strengths of this study include a relatively large population sample of prepubertal children, the assessment of dietary factors by food records that were checked individually by a clinical nutritionist, the objective assessment of physical activity and sedentary behavior including intensity classification by individually calibrated combined heart rate and movement monitoring, the detailed assessment of different types of physical activity and sedentary behavior by a questionnaire, and the accurate assessment of body fat percentage by DXA. A limitation of the study is that, although guided carefully, food consumption was assessed using food records reported by the parents, which may have caused misreporting.
Our study provides new information on the associations between lifestyle factors and adrenarche, which typically occurs before the onset of puberty (39). Although adiposity plays a major role in the development of adrenarche, our results show that dietary factors, physical activity, and sedentary behavior have weak to moderate associations with serum DHEAS and IGF-1 levels in prepubertal children. The associations of dietary factors and serum DHEAS and IGF-1 levels tend to be partly independent of body fat percentage whereas those of physical activity and sedentary behavior seem to be explained by adiposity. However, based on this and several previous studies, the most effective way to control adrenal androgen secretion is to prevent overweight.
Acknowledgments
We thank all children and their families who participated in this study. We are also gratefully indebted to the PANIC study research team members for their skillful contribution in performing the study. We also thank Ms Leila Antikainen for biochemical analyses. This work has been financially supported by grants from Ministry of Social Affairs and Health of Finland, Ministry of Education and Culture of Finland, Finnish Innovation Fund Sitra, Social Insurance Institution of Finland, Finnish Cultural Foundation, Juho Vainio Foundation, Yrjö Jahnsson Foundation, Foundation for Pediatric Research, Paulo Foundation, Paavo Nurmi Foundation, Diabetes Research Foundation, Finnish Foundation for Cardiovascular Research, Research Committee of Kuopio University Hospital Catchment Area (State Research Funding) and Kuopio University Hospital (EVO funding number 5031343), Finnish Medical Foundation, Päivikki and Sakari Sohlberg Foundation, Rauha and Jalmari Ahokas Foundation, and the city of Kuopio.
All authors have made substantial contribution to a) conception and design, acquisition of data, or analysis and interpretation of data; b) drafting the article or revising it critically for important intellectual content; and c) given final approval of the version to be published.
Footnotes
Clinical Trial Registration Number: NCT01803776, www.clinicaltrials.gov
Conflict of interest statement: Authors report no conflicts of interest related to the study.
References
- 1.Auchus RJ, Rainey WE. Adrenarche - physiology, biochemistry and human disease. Clin Endocrinol (Oxf) 2004;60(3):288–296. doi: 10.1046/j.1365-2265.2003.01858.x. [DOI] [PubMed] [Google Scholar]
- 2.Utriainen P, Laakso S, Liimatta J, Jääskeläinen J, Voutilainen R. Premature adrenarche - a common condition with variable presentation. Horm Res Paediatr. 2015;83(4):221–231. doi: 10.1159/000369458. [DOI] [PubMed] [Google Scholar]
- 3.Ong KK, Potau N, Petry CJ, et al. Opposing influences of prenatal and postnatal weight gain on adrenarche in normal boys and girls. J Clin Endocrinol Metab. 2004;89(6):2647–2651. doi: 10.1210/jc.2003-031848. [DOI] [PubMed] [Google Scholar]
- 4.Thankamony A, Ong KK, Ahmed ML, Ness AR, Holly JM, Dunger DB. Higher levels of IGF-I and adrenal androgens at age 8 years are associated with earlier age at menarche in girls. J Clin Endocrinol Metab. 2012;97(5):E786–90. doi: 10.1210/jc.2011-3261. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Pratt JH, Manatunga AK, Li W. Familial influences on the adrenal androgen excretion rate during the adrenarche. Metabolism. 1994;43(2):186–189. doi: 10.1016/0026-0495(94)90243-7. [DOI] [PubMed] [Google Scholar]
- 6.Utriainen P, Jääskeläinen J, Romppanen J, Voutilainen R. Childhood metabolic syndrome and its components in premature adrenarche. J Clin Endocrinol Metab. 2007;92(11):4282–4285. doi: 10.1210/jc.2006-2412. [DOI] [PubMed] [Google Scholar]
- 7.Shi L, Wudy SA, Buyken AE, Hartmann MF, Remer T. Body fat and animal protein intakes are associated with adrenal androgen secretion in children. Am J Clin Nutr. 2009;90(5):1321–1328. doi: 10.3945/ajcn.2009.27964. [DOI] [PubMed] [Google Scholar]
- 8.Corvalán C, Uauy R, Mericq V. Obesity is positively associated with dehydroepiandrosterone sulfate concentrations at 7 y in Chilean children of normal birth weight. Am J Clin Nutr. 2013;97(2):318–325. doi: 10.3945/ajcn.112.037325. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Garnett SP, Hogler W, Blades B, et al. Relation between hormones and body composition, including bone, in prepubertal children. Am J Clin Nutr. 2004;80(4):966–972. doi: 10.1093/ajcn/80.4.966. [DOI] [PubMed] [Google Scholar]
- 10.Remer T, Manz F. Role of nutritional status in the regulation of adrenarche. J Clin Endocrinol Metab. 1999;84(11):3936–3944. doi: 10.1210/jcem.84.11.6093. [DOI] [PubMed] [Google Scholar]
- 11.Mäntyselka A, Jääskeläinen J, Lindi V, et al. The presentation of adrenarche is sexually dimorphic and modified by body adiposity. J Clin Endocrinol Metab. 2014;99(10):3889–3894. doi: 10.1210/jc.2014-2049. [DOI] [PubMed] [Google Scholar]
- 12.Utriainen P, Voutilainen R, Jääskeläinen J. Girls with premature adrenarche have accelerated early childhood growth. J Pediatr. 2009;154(6):882–887. doi: 10.1016/j.jpeds.2008.12.038. [DOI] [PubMed] [Google Scholar]
- 13.Voutilainen R, Miller WL. Coordinate tropic hormone regulation of mRNAs for insulin-like growth factor II and the cholesterol side-chain-cleavage enzyme, P450scc [corrected], in human steroidogenic tissues. Proc Natl Acad Sci U S A. 1987;84(6):1590–1594. doi: 10.1073/pnas.84.6.1590. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Baquedano MS, Berensztein E, Saraco N, et al. Expression of the IGF system in human adrenal tissues from early infancy to late puberty: Implications for the development of adrenarche. Pediatr Res. 2005;58(3):451–458. doi: 10.1203/01.PDR.0000179392.59060.93. [DOI] [PubMed] [Google Scholar]
- 15.l'Allemand D, Penhoat A, Lebrethon MC, et al. Insulin-like growth factors enhance steroidogenic enzyme and corticotropin receptor messenger ribonucleic acid levels and corticotropin steroidogenic responsiveness in cultured human adrenocortical cells. J Clin Endocrinol Metab. 1996;81(11):3892–3897. doi: 10.1210/jcem.81.11.8923834. [DOI] [PubMed] [Google Scholar]
- 16.de Ridder CM, Thijssen JH, Van't Veer P, et al. Dietary habits, sexual maturation, and plasma hormones in pubertal girls: A longitudinal study. Am J Clin Nutr. 1991;54(5):805–813. doi: 10.1093/ajcn/54.5.805. [DOI] [PubMed] [Google Scholar]
- 17.Porter JR, Svec F. DHEA diminishes fat food intake in lean and obese zucker rats. Ann N Y Acad Sci. 1995;774:329–331. doi: 10.1111/j.1749-6632.1995.tb17400.x-i1. [DOI] [PubMed] [Google Scholar]
- 18.Gillen G, Porter JR, Svec F. Synergistic anorectic effect of dehydroepiandrosterone and d-fenfluramine on the obese zucker rat. Physiol Behav. 1999;67(2):173–179. doi: 10.1016/s0031-9384(99)00057-8. [DOI] [PubMed] [Google Scholar]
- 19.Rogers IS, Gunnell D, Emmett PM, Glynn LR, Dunger DB, Holly JM. Cross-sectional associations of diet and insulin-like growth factor levels in 7- to 8-year-old children. Cancer Epidemiol Biomarkers Prev. 2005;14(1):204–212. [PubMed] [Google Scholar]
- 20.Kerver JM, Gardiner JC, Dorgan JF, Rosen CJ, Velie EM. Dietary predictors of the insulin-like growth factor system in adolescent females: Results from the dietary intervention study in children (DISC) Am J Clin Nutr. 2010;91(3):643–650. doi: 10.3945/ajcn.2009.28205. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Hoppe C, Udam TR, Lauritzen L, Molgaard C, Juul A, Michaelsen KF. Animal protein intake, serum insulin-like growth factor I, and growth in healthy 2.5-y-old danish children. Am J Clin Nutr. 2004;80(2):447–452. doi: 10.1093/ajcn/80.2.447. [DOI] [PubMed] [Google Scholar]
- 22.Hoppe C, Molgaard C, Juul A, Michaelsen KF. High intakes of skimmed milk, but not meat, increase serum IGF-I and IGFBP-3 in eight-year-old boys. Eur J Clin Nutr. 2004;58(9):1211–1216. doi: 10.1038/sj.ejcn.1601948. [DOI] [PubMed] [Google Scholar]
- 23.Wada K, Nakamura K, Tamai Y, et al. Associations of birth weight and physical activity with sex steroids in preschool japanese children. Cancer Causes Control. 2012;23(2):231–238. doi: 10.1007/s10552-011-9870-0. [DOI] [PubMed] [Google Scholar]
- 24.Berg U, Bang P. Exercise and circulating insulin-like growth factor I. Horm Res. 2004;1(62 Suppl):50–58. doi: 10.1159/000080759. [DOI] [PubMed] [Google Scholar]
- 25.Eliakim A, Scheett TP, Newcomb R, Mohan S, Cooper DM. Fitness, training, and the growth hormone-->insulin-like growth factor I axis in prepubertal girls. J Clin Endocrinol Metab. 2001;86(6):2797–2802. doi: 10.1210/jcem.86.6.7560. [DOI] [PubMed] [Google Scholar]
- 26.Eliakim A, Nemet D. The endocrine response to exercise and training in young athletes. Pediatr Exerc Sci. 2013;25(4):605–615. doi: 10.1123/pes.25.4.605. [DOI] [PubMed] [Google Scholar]
- 27.Eloranta AM, Lindi V, Schwab U, et al. Dietary factors and their associations with socioeconomic background in finnish girls and boys 6-8 years of age: The PANIC study. Eur J Clin Nutr. 2011;65(11):1211–1218. doi: 10.1038/ejcn.2011.113. [DOI] [PubMed] [Google Scholar]
- 28.Rosenfield RL. Clinical review: Identifying children at risk for polycystic ovary syndrome. J Clin Endocrinol Metab. 2007;92(3):787–796. doi: 10.1210/jc.2006-2012. [DOI] [PubMed] [Google Scholar]
- 29.Saari A, Sankilampi U, Hannila ML, Kiviniemi V, Kesseli K, Dunkel L. New finnish growth references for children and adolescents aged 0 to 20 years: Length/height-for-age, weight-for-length/height, and body mass index-for-age. Ann Med. 2011;43(3):235–248. doi: 10.3109/07853890.2010.515603. [DOI] [PubMed] [Google Scholar]
- 30.Rastas M, Seppänen R, Knuts LR, et al. Nutrient composition of foods. Helsinki: Publications of the social insurance institution; 1997. [Google Scholar]
- 31.Brage S, Brage N, Franks PW, Ekelund U, Wareham NJ. Reliability and validity of the combined heart rate and movement sensor actiheart. Eur J Clin Nutr. 2005;59(4):561–570. doi: 10.1038/sj.ejcn.1602118. [DOI] [PubMed] [Google Scholar]
- 32.Väistö J, Eloranta AM, Viitasalo A, et al. Physical activity and sedentary behaviour in relation to cardiometabolic risk in children: Cross-sectional findings from the physical activity and nutrition in children (PANIC) study. Int J Behav Nutr Phys Act. 2014;11(1):55. doi: 10.1186/1479-5868-11-55. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Guercio G, Rivarola MA, Chaler E, Maceiras M, Belgorosky A. Relationship between the GH/IGF-I axis, insulin sensitivity, and adrenal androgens in normal prepubertal and pubertal boys. J Clin Endocrinol Metab. 2002;87(3):1162–1169. doi: 10.1210/jcem.87.3.8330. [DOI] [PubMed] [Google Scholar]
- 34.Guercio G, Rivarola MA, Chaler E, Maceiras M, Belgorosky A. Relationship between the growth hormone/insulin-like growth factor-I axis, insulin sensitivity, and adrenal androgens in normal prepubertal and pubertal girls. J Clin Endocrinol Metab. 2003;88(3):1389–1393. doi: 10.1210/jc.2002-020979. [DOI] [PubMed] [Google Scholar]
- 35.Courant F, Aksglaede L, Antignac JP, et al. Assessment of circulating sex steroid levels in prepubertal and pubertal boys and girls by a novel ultrasensitive gas chromatography-tandem mass spectrometry method. J Clin Endocrinol Metab. 2010;95(1):82–92. doi: 10.1210/jc.2009-1140. [DOI] [PubMed] [Google Scholar]
- 36.Veldhuis JD, Metzger DL, Martha PM, Jr, et al. Estrogen and testosterone, but not a nonaromatizable androgen, direct network integration of the hypothalamo-somatotrope (growth hormone)-insulin-like growth factor I axis in the human: Evidence from pubertal pathophysiology and sex-steroid hormone replacement. J Clin Endocrinol Metab. 1997;82(10):3414–3420. doi: 10.1210/jcem.82.10.4317. [DOI] [PubMed] [Google Scholar]
- 37.Collomp K, Buisson C, Lasne F, Collomp R. DHEA, physical exercise and doping. J Steroid Biochem Mol Biol. 2015;145:206–212. doi: 10.1016/j.jsbmb.2014.03.005. [DOI] [PubMed] [Google Scholar]
- 38.Collings PJ, Westgate K, Väistö J, et al. Cross-sectional associations of objectively-measured physical activity and sedentary time with body composition and cardiorespiratory fitness in mid-childhood: The PANIC study. Sports Med. 2016 doi: 10.1007/s40279-016-0606-x. e-pub ahead of print 24 August 2016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Biro FM, Pinney SM, Huang B, Baker ER, Walt Chandler D, Dorn LD. Hormone changes in peripubertal girls. J Clin Endocrinol Metab. 2014;99(10):3829–3835. doi: 10.1210/jc.2013-4528. [DOI] [PMC free article] [PubMed] [Google Scholar]