Skip to main content
NCBI home page
Bone Reports logoLink to Bone Reports
. 2021 May 4;14:101087. doi: 10.1016/j.bonr.2021.101087

Sex differences in the relationship between body composition and biomarkers of bone and fat metabolism in obese boys and girls

Rapheeporn Khwanchuea 1,⁎, Chuchard Punsawad 1
PMCID: PMC8121990  PMID: 34026951

Abstract

Whether a body mass derived from extremes of body weight is beneficial to bone remains controversial. When fat accumulation reaches excessive levels and induces changes in hormonal factors and adipokines, it may affect bone accrual during growth. This study evaluated the relationships between body composition and key biomarkers in relation to bone and fat metabolism in obese Thai boys and girls. Subjects aged 12–14 years were grouped by body mass index (BMI) and percentage of body fat (%Fat). Body composition and heel bone Z-score and speed of sound (SOS) were assessed by bioelectrical impedance analysis and calcaneus bone densitometry, respectively. Serum osteocalcin (OC), adiponectin, leptin, insulin, and 25 hydroxyvitamin D (25(OH)D) were measured by ELISA. Their correlations were analyzed and compared between sexes. The results showed that the obese groups had no differences in mean BMIs and body composition, except that boys had more muscle mass than girls. Boys had lower serum OC and leptin levels than girls. Positive correlations of leptin with %Fat and FM were found in both sexes, while positive associations of %Fat with OC and insulin were found only in boys. Bone Z-score and SOS positively correlated with OC in boys but negatively correlated with 25(OH)D in girls. When classifying the obese group using %Fat ≥25, the positive correlations between %Fat and insulin and the negative associations between %Fat and adiponectin in girls were more pronounced. These results suggest that the associations of body fat and bone parameters with OC, adiponectin, 25(OH)D, and insulin were sex-specific, with greater clarity when %Fat was used instead of BMI to classify obesity.

Abbreviations: BW, body weight; BMC, bone mineral content; %Fat, percentage of body fat; aBMD, areal bone mineral density; BMD, bone mineral density; OC, osteocalcin; IR, insulin resistance; BMI, body mass index; 25(OH)D, 25-hydroxyvitamin D; FMI, fat mass index; FFMI, free fat mass index; FM, fat mass; MM, muscle mass; FFM, free fat mass; SOS, speed of sound; ELISA, enzyme-linked immunosorbent assay

Keywords: Body fat percentage, Osteocalcin, Adiponectin, Leptin, 25(OH)D, Adolescents

Highlights

  • •

    The associations of adiponectin, OC, 25(OH)D and insulin with %Fat and bone mass were sex-specific.

  • •

    Excessive %Fat might be risk for bone mass due to the negative relationships with 25(OH)D and OC.

  • •

    A negative effect of body fat but a favorable effect of muscle mass on bone mass in girls.

1. Introduction

The interaction between obesity and bone metabolism is complex. Obesity is traditionally viewed to be beneficial to bone health, but whether the mass derived from extremes of BW or excessive fat accumulation is beneficial to bone remains controversial (Cao, 2011). A previous study demonstrated lower whole-body bone mineral content (BMC) in obese children with a percentage of body fat (%Fat) > 30 than in leaner children (%Fat <25) (Júnior et al., 2013). Recently, a study found a negative association between truncal fat mass and areal bone mineral density (aBMD) Z-score in children with levels of fat mass in the upper 15th percentile (Rokoff et al., 2019). However, evidence has shown that bone parameters increase progressively with age, specifically in gender and stage of puberty, and a rapid increase in bone mineral density (BMD) and BMC was observed at earlier ages in girls (Nakavachara et al., 2014). A recent review revealed a significantly high bone mass as estimated by whole-body BMD in girls with obesity but not in boys (Fintini et al., 2020).

Several studies have indicated that fat-induced changes in hormonal factors and cytokines, particularly in relation to adipokines, play a pivotal role in disturbing bone accrual during growth (Dimitri et al., 2012; Dimitri, 2018; Faienza et al., 2019). A previous study reported that adolescents with obesity and higher serum leptin concentrations have higher BMD than lean adolescents despite lower serum vitamin D and physical activity levels (Maggio et al., 2014). Recently, a study showed that positive relationships between serum leptin levels at age 11 years and bone mineral parameters at age 14 years are independent of body fat percentage in school-aged healthy children (Kouda et al., 2019). Conversely, a negative association of leptin with BMD was reported to be independent of body size but modulated by sex, particularly in boys (Armaiz-Flores et al., 2017).

In human obesity, hyperleptinemia is associated with reductions in bone formation biomarkers, especially osteocalcin (OC) (Fernandes et al., 2017). OC has endocrine functions in the regulation of energy expenditure and glucose by stimulating insulin secretion in adipose tissue (O'Connor and Edel, 2017). Serum OC was demonstrated to be lower in obese children due to insulin resistance (IR) and leptin (Reinehr and Roth, 2010), and IR has an unfavorable effect on BMC that is more obvious in males than in females (Lee, 2013; Park et al., 2016). In addition, OC signals in adipocytes secrete adiponectin, where concentrations of adiponectin are inversely related to the degree of adiposity. A previous study indicated that boys have lower adiponectin concentrations than girls, meaning that adiponectin decreased for every unit increase in body mass index (BMI) Z-score (Punthakee et al., 2006). Moreover, evidence has reported that 25-hydroxyvitamin D (25(OH)D) is positively correlated with OC and adiponectin and is negatively associated with BMI, body weight (BW), and waist circumference among adolescents aged 14–18 years (Giudici et al., 2017).

In obesity, BMI and %Fat are widely used to assess adiposity. Among children with a BMI-for-age ≥ 85th percentile, BMI levels are strongly associated with fat mass index (FMI); in contrast, among children with a BMI-for-age < 50th percentile, BMI levels are more strongly associated with free fat mass index (FFMI) than with FMI (Freedman et al., 2005). After 12 years of age, girls had slightly higher BMI levels than boys because the sex difference in FMI is greater than that for FFMI, and FMI levels decrease after 12 years of age among boys. Therefore, in this study, it was interesting to use BMI-for-age and excessive %Fat to classify obesity in subjects aged approximately 13–14 years.

However, in Thai early adolescents, there have been few studies about the associations between body composition (focusing on %Fat, fat mass (FM) and muscle mass (MM)) and biomarkers relating to bone and fat metabolism in obesity. Therefore, this study aimed to evaluate the relationships between body composition and key biomarkers of bone and fat metabolism, including OC, adiponectin, leptin, 25(OH)D, and insulin, in obese Thai boys and girls.

2. Materials and methods

2.1. Study subjects

Eighty-four subjects aged 12–14 years were included from an urban secondary school in southern Thailand with BMI-for-age ≥95th percentile and <50th percentile and with the exclusion criteria of any chronic or bone diseases; other conditions such as asthma, allergies, or gastritis; and use of steroids or anticonvulsant drugs. Subjects were divided by sex (boys and girls) and grouped into obese and control groups according to BMI-for-age ≥95th percentile and <50th percentile, respectively (de Onis et al., 2007; World Health Organization. Regional Office for the Western, 2000). In addition to BMI, obese subjects were regrouped using %Fat ≥25 (or excessive %Fat), which was calculated individually during body composition measurement. Subjects were asked about their lifestyle habits: furthermore, they did not exercise regularly; during the day they spent most of their time sitting in class and ate 3 meals a day with snacks and no calcium supplementation.

This study was approved by the ethics committee on Human Rights Related to Research Involving Human Subjects, Walailak University, Thailand. A consent form was obtained from all subjects or their legal representative before enrollment.

2.2. Anthropometric and body composition measurements

Body weight (kg), BMI (kg/m2), and body composition, including %Fat, FM (kg), free fat mass (FFM, kg), and MM (kg), were analyzed by bioelectrical impedance analysis (Talma et al., 2013; Dobroch et al., 2018) using a TANITA SC-330ST series body composition analyzer (Tanita Corporation, Tokyo, Japan). To increase the accuracy of measurements, subjects were wearing light clothes and were not wearing shoes. Apart from clothing, 0.5 kg of weight was deducted due to the level of hydration or the presence of edema together with the daily variability, which also affected the obtained values. Standing heights (m) were measured without shoes using a locally made stadiometer and were recorded to the nearest 0.1 cm.

2.3. Measurement of bone mass

Bone Z-score and speed of sound (SOS, m/s) were assessed by calcaneus quantitative ultrasound at the left heel using a calcaneus bone densitometer (AOS-100SA, Aloka Co., Ltd., Tokyo, Japan). Calcaneal quantitative ultrasound is able to reflect bone quality, and ultrasound velocity is influenced by structural bone variables that are dependent on bone density (Chin and Ima-Nirwana, 2013; Baroncelli, 2008). This measurement was performed with subjects sitting in a chair and bending their ankle to 90° (midway between plantar flexion and dorsal flexion), and SOS was measured using an ultrasound wave passing through the calcaneus. A normal bone mass was accepted at a Z-score of more than −1.0 (Schraders et al., 2019).

2.4. Measurements of biomarkers for bone and fat metabolisms by enzyme-linked immunosorbent assay (ELISA)

After an overnight fast, blood sample was collected to clotted blood. The clotted blood was centrifuged at 2000 rpm for 10 min, and the serum was harvested and stored in an aliquoted state at −70 °C until used for the measurement of OC (ng/mL), adiponectin (ng/mL), leptin (ng/mL), insulin (mU/L), and 25(OH)D (ng/mL). Commercial ELISA kits were used for the measurement of OC (R&D Systems, Inc., Minneapolis, MN, USA, sensitivity to 0.898 ng/mL and intra-assay coefficient of variability <10%), adiponectin (Abcam, Cambridge, UK, sensitivity to 25 pg/mL and intra-assay coefficient of variability <10%), leptin (R&D Systems, Inc., Minneapolis, MN, USA, sensitivity to 7.8 pg/mL and intra-assay coefficient of variability <10%), insulin (Abcam, Cambridge, UK, sensitivity to 4 mU/L and intra-assay coefficient of variability <10%), and 25(OH)D (DRG Diagnostics, Frauenbergstrasse, Germany, sensitivity to 2.89 ng/mL and intra-assay coefficient of variability <10%), according to the manufacturers' instructions. In brief, standard or diluted serum samples were added in coated microplate and incubated for 1 h at room temperature. The plates were then washed and incubated with antibody cocktail as the mixture of capture and detector antibodies for 2 h at room temperature. After washing, to detect the reaction of antigen-antibody complex, 3,3′,5,5′-tetramethylbenzidine (TMB) substrate was added to each well and incubated for 10 min. Finally, to terminate the peroxidase reaction, stop solution was added to each well, and measured the optical density (OD) at 450 nm. All assays were performed in duplicate.

2.5. Statistical analysis

The results were presented as mean ± SD. Data analysis was performed using IBM SPSS statistics version 22.0 software license authorization wizard. Differences in parameters of bone mass, anthropometrics, body composition, and biomarkers among control groups and obese groups were compared by independent-samples t-test. The correlations between those parameters were calculated by Pearson's correlation coefficient. The p-values less than 0.05, 0.01 or 0.001 were considered statistically significant.

3. Results

Eighty-four subjects (mean ages = 13.31 ± 0.69) were grouped by sex and classified using BMI for age. As shown in Table 1, in the obese groups, neither sex had differences in mean BMIs, %Fat, FM, FFM, adiponectin, 25(OH)D, or insulin. Obese boys had higher BW and MM but lower serum OC and leptin than obese girls. Both sexes had normal calcaneus bone mass, and girls had higher Z-scores than boys, with no difference in SOS. In addition, BMI of obese boys and girls positively correlated with %Fat (r = 0.924, p < 0.001; r = 0.609, p < 0.001) and FM (r = 0.943, p < 0.001; r = 0.814, p < 0.001); at the same time, positive associations of BMI with FFM (r = 0.438, p = 0.016) and MM (r = 0.438, p = 0.016) were found only in obese girls (data not shown in the table). When compared to control boys, obese boys had higher BW, BMI, %Fat, FM, leptin, and insulin but lower adiponectin and 25(OH)D (p < 0.05), whereas bone Z-score, SOS, FFM, MM, and OC were not different. In girls, the obese group had higher values for mean bone Z-score and all parameters of body composition, leptin, and insulin but lower adiponectin and 25(OH)D than the control group, while SOS and serum OC were not different.

Table 1.

Comparisons of bone parameters, body composition, and biomarkers between control and obese groups of boy and girl, and between obese boys and girls classified by BMI.

Boy
 Girl
 Boy and girl
Control group N = 14 Obese group N = 28 p-value Control group N = 13 Obese group N = 29 p-value Obese groups p-value
Bone Z-score 0.02 ± 1.23 0.66 ± 0.84 0.054 0.88 ± 2.53 3.08 ± 3.22 0.035 <0.001
SOS (m/s) 1557 ± 13.72 1560.61 ± 14.02 0.480 1561.23 ± 22.66 1570.10 ± 32.45 0.378 0.160
BW (kg) 56.27 ± 7.04 78.43 ± 11.36 <0.001 53.61 ± 5.53 72.83 ± 8.21 <0.001 0.037
BMI (kg/m2) 20.95 ± 2.19 28.96 ± 3.42 <0.001 21.69 ± 1.68 28.26 ± 2.68 <0.001 0.394
%Fat 20.48 ± 7.45 39.54 ± 7.88 <0.001 27.59 ± 6.90 39.03 ± 4.46 <0.001 0.765
FM (kg) 11.79 ± 5.03 31.63 ± 10.54 <0.001 15.04 ± 4.82 28.59 ± 5.94 <0.001 0.183
FFM (kg) 44.48 ± 4.98 45.96 ± 6.17 0.440 38.57 ± 2.51 44.24 ± 4.35 <0.001 0.228
MM (kg) 42.12 ± 4.68 44.30 ± 4.69 0.163 36.23 ± 2.30 41.43 ± 3.99 <0.001 0.016



Boy
 Girl
 Boy and girl
Control group N = 14 Obese group N = 23 p-value Control group N = 12 Obese group N = 28 p-value Obese groups p-value
Adiponectin (ng/mL) 2.66 ± 0.90 2.05 ± 0.72 0.029 2.80 ± 0.97 2.00 ± 0.61 0.018 0.795
OC (ng/mL) 10.20 ± 1.00 10.67 ± 1.58 0.328 13.14 ± 1.52 12.59 ± 1.89 0.375 <0.001
25(OH)D (ng/mL) 27.17 ± 3.10 24.67 ± 2.43 0.010 26.97 ± 1.52 23.99 ± 3.11 0.003 0.396
Leptin (ng/mL) 0.55 ± 0.55 1.22 ± 0.67 0.003 1.26 ± 0.51 1.86 ± 0.45 0.001 <0.001
Insulin (mU/L) 7.61 ± 5.96 20.42 ± 20.58 0.030 9.78 ± 6.99 21.51 ± 17.65 0.005 0.840
Open in a new tab

Values are presented as mean ± SD.

Table 2 shows that in obese boys, bone parameters (Z-score and SOS) were positively correlated with OC, and BMI, %Fat, and FM were positively associated with leptin and insulin, while BW was positively associated with only leptin. In obese girls, bone Z-score and SOS were negatively correlated with 25(OH)D, whereas BW, BMI, %Fat, and FM were positively correlated with leptin. There was no association between adiponectin and parameters of bone and body composition. In addition, the data are not shown in the table. There was a positive correlation between leptin and insulin in obese boys (r = 0.537, p = 0.005) and weakly negative associations of 25(OH)D with leptin and insulin in obese girls (r = −0.372, p = 0.043; r = −0.413, p = 0.023).

Table 2.

Correlation coefficients (r) of bone parameters and body composition with biomarkers in obese boys and obese girls classified by BMI.

Obese boys N = 23
Adiponectin (ng/mL)
 OC (ng/mL)
 25(OH)D (ng/mL)
 Leptin (ng/mL)
 Insulin (mU/L)
r p-value r p-value r p-value r p-value r p-value
Bone Z-score −0.124 0.572 0.415 0.049 −0.007 0.976 0.112 0.610 −0.05 0.820
SOS (m/s) 0.025 0.909 0.700 <0.001 0.102 0.643 0.053 0.809 −0.041 0.854
BW (kg) −0.178 0.415 0.281 0.194 −0.039 0.860 0.489 0.018 0.379 0.075
BMI (kg/m2) −0.102 0.642 0.406 0.055 0.077 0.726 0.652 0.001 0.536 0.008
%Fat −0.081 0.715 0.455 0.029 0.230 0.291 0.657 0.001 0.421 0.045
FM (kg) −0.109 0.620 0.399 0.059 0.099 0.655 0.648 0.001 0.458 0.028
FFM (kg) −0.181 0.408 0.045 0.840 −0.270 0.212 0.016 0.942 −0.013 0.952
MM (kg) −0.195 0.374 −0.13 0.555 −0.277 0.200 −0.136 0.537 −0.021 0.923



Obese girls N = 28
Adiponectin (ng/mL)
 OC (ng/mL)
 25(OH)D (ng/mL)
 Leptin (ng/mL)
 Insulin (mU/L)
r p-value r p-value r p-value r p-value r p-value
Bone Z-score −0.012 0.952 0.036 0.857 −0.482 0.009 0.248 0.204 0.348 0.069
SOS (m/s) −0.040 0.838 0.013 0.948 −0.673 <0.001 0.348 0.070 0.354 0.065
BW (kg) 0.232 0.235 −0.019 0.923 −0.171 0.384 0.449 0.017 0.306 0.113
BMI (kg/m2) 0.069 0.727 0.069 0.728 −0.235 0.229 0.535 0.003 0.265 0.173
%Fat 0.104 0.599 0.165 0.400 −0.143 0.468 0.619 <0.001 0.250 0.200
FM (kg) 0.204 0.298 0.086 0.662 −0.174 0.375 0.618 <0.001 0.318 0.100
FFM (kg) 0.157 0.426 −0.155 0.431 −0.082 0.677 −0.006 0.977 0.140 0.479
MM (kg) 0.156 0.429 −0.156 0.429 −0.084 0.669 −0.005 0.980 0.141 0.475
Open in a new tab

When regrouping obese subjects using %Fat ≥25, Table 3 shows that boys and girls had no differences in SOS, BMI, %Fat, adiponectin, 25(OH)D, and insulin; however, boys had higher BW, FM, FFM, and MM but lower bone Z-score, OC, and leptin than girls. Table 4 shows that in boys, in addition to bone Z-score and SOS, OC was positively correlated with %Fat. Similar to the results shown in Table 2, BMI, %Fat, and FM were positively associated with leptin and insulin, but BW was positively associated with only leptin in boys. The results in girls showed that parameters reflecting obesity (BW, BMI, %Fat, and FM) were negatively correlated with adiponectin and 25(OH)D; at the same time, bone Z-score and SOS were negatively associated with 25(OH)D and insulin. Moreover, leptin was positively associated with all parameters of body composition, and insulin was positively correlated with bone Z-score and SOS, including BW, BMI, %Fat, and FM.

Table 3.

Comparisons of bone parameters, body composition, and biomarkers between boys and girls with %Fat ≥25.

Boys N = 32 Girls N = 40 p-value
Bone Z-score 0.71 ± 0.95 2.54 ± 3.15 0.002
SOS (m/s) 1561.25 ± 14.10 1567.10 ± 30.22 0.282
BW (kg) 76.32 ± 12.12 67.88 ± 11.01 0.003
BMI (kg/m2) 28.25 ± 3.72 26.52 ± 3.74 0.055
%Fat 38.30 ± 8.19 36.51 ± 5.89 0.284
FM (kg) 29.94 ± 10.85 25.26 ± 7.63 0.035
FFM (kg) 45.64 ± 6.07 42.62 ± 4.71 0.020
MM (kg) 43.90 ± 4.79 39.95 ± 4.33 <0.001



Boys N = 27 Girls N = 38 p-value
Adiponectin (ng/mL) 2.13 ± 0.78 2.21 ± 0.82 0.665
OC (ng/mL) 10.63 ± 1.52 12.84 ± 1.76 <0.001
25(OH)D (ng/mL) 25.00 ± 2.48 24.73 ± 3.04 0.703
Leptin (ng/mL) 1.14 ± 0.65 1.70 ± 0.53 <0.001
Insulin (mU/L) 18.88 ± 19.52 18.31 ± 16.46 0.899
Open in a new tab

Values are presented as mean ± SD.

Table 4.

Correlation coefficients (r) of bone Z-score and body composition with biomarkers in boys and girls with %Fat ≥25.

Boys N = 27
Adiponectin (ng/mL)
 OC (ng/mL)
 25(OH)D (ng/mL)
 Leptin (ng/mL)
 Insulin (mU/L)
r p-value r p-value r p-value r p-value r p-value
Bone Z-score −0.211 0.291 0.412 0.033 0.090 0.656 0.021 0.918 −0.092 0.648
SOS (m/s) −0.026 0.898 0.580 0.002 0.072 0.723 0.019 0.925 −0.035 0.861
BW (kg) −0.262 0.187 0.291 0.141 −0.159 0.428 0.526 0.005 0.399 0.039
BMI (kg/m2) −0.234 0.240 0.350 0.074 −0.128 0.525 0.665 0.000 0.536 0.004
%Fat −0.209 0.295 0.388 0.045 0.039 0.848 0.694 0.000 0.472 0.013
FM (kg) −0.214 0.284 0.368 0.059 −0.057 0.779 0.680 0.000 0.492 0.009
FFM (kg) −0.176 0.381 0.106 0.597 −0.243 0.222 0.031 0.878 −0.017 0.933
MM (kg) −0.203 0.311 −0.025 0.902 −0.262 0.187 −0.074 0.713 −0.011 0.958



Girls N = 38
Adiponectin (ng/mL)
 OC (ng/mL)
 25(OH)D (ng/mL)
 Leptin (ng/mL)
 Insulin (mU/L)
r p-value r p-value r p-value r p-value r p-value
Bone Z-score −0.142 0.396 0.002 0.990 −0.498 0.001 0.217 0.190 0.412 0.010
SOS (m/s) −0.104 0.534 −0.015 0.930 −0.648 <0.001 0.231 0.162 0.407 0.011
BW (kg) −0.323 0.048 −0.170 0.309 −0.411 0.010 0.668 0.000 0.429 0.007
BMI (kg/m2) −0.399 0.013 −0.133 0.425 −0.451 0.004 0.688 0.000 0.427 0.007
%Fat −0.406 0.012 −0.063 0.709 −0.404 0.012 0.724 0.000 0.424 0.008
FM (kg) −0.327 0.045 −0.108 0.518 −0.417 0.009 0.735 0.000 0.451 0.004
FFM (kg) −0.221 0.182 −0.220 0.184 −0.281 0.088 0.360 0.026 0.267 0.105
MM (kg) −0.221 0.182 −0.221 0.183 −0.283 0.086 0.360 0.026 0.268 0.104
Open in a new tab

4. Discussion

This study included obese subjects aged 12–14 years using BMI-for-age ≥ 95th percentile as a determinant, as the mean BMIs of boys and girls were not significantly different and associations between BMI and body fat (%Fat and FM) in both sexes were similar, although height, BW, and MM were greater in boys than in girls. The BW of obese boys might include MM, indicating body proportion and fat distribution changes during the pubertal years; males gain greater amounts of fat-free mass and skeletal mass, while females acquire significantly more fat mass (Loomba-Albrecht and Styne, 2009). However, we found a small association between BMI and MM in obese girls, and the bone Z-score of obese girls was higher than that of obese boys. These data may be explained by a favorable effect of MM on calcaneus bone mass in obese girls and their earlier puberty that sex steroids act in concert with growth hormone to increase bone density (Nakavachara et al., 2014; Jaruratanasirikul et al., 2014).

When regrouping obese subjects using %Fat ≥25, the number of obese subjects classified by excessive %Fat was greater than that grouped by BMI ≥ 95th percentile. The mean BMI of the obese girls with %Fat ≥25 was lower than that with BMI ≥ 95th percentile. This may be explained according to a report which described that %Fat is a better indicator of weight status than BMI, mainly among girls (Oliosa et al., 2019). Body fat mass in subjects with %Fat ≥25 was more frequently detected in boys than in girls in addition to FFM and MM. These findings suggest that the body composition of obese subjects varies by sex and that BMIs of approximately 28–29 might reflect excessive body fat more accurately in girls than in boys.

The results of biomarker analyses in both sexes showed high serum leptin in the obese groups and positive associations of leptin with all parameters for obesity (BW, BMI, %Fat, and FM), consistent with the previous finding that leptin plays a key role as a biomarker for childhood obesity and can help in the prediction of weight gain in obese children (Alamri et al., 2017). This study found that relationships between leptin and body fat were stronger in girls than in boys; OC and leptin were lower in boys than in girls despite the lack of differences in %Fat and FM between sexes; relationships between OC and %Fat and heel bone Z-score were found only in boys. These results suggest that OC may be a promising link between obesity and IR (Lee, 2013). When compared with normal-weight subjects, serum leptin and insulin levels in obese subjects were higher. Although obese subjects were not diagnosed with IR, this result should be considered with a previous report indicating that the unfavorable effect of IR on bone is more obvious in males aged 13–19 years or nonoverweight males than in females (Park et al., 2016). In our results, OC and heel bone Z-scores were lower in boys than in girls, and OC in obese boys positively correlated with heel bone Z-score and %Fat. These findings indicated sex-specific correlations of OC with insulin, leptin, and bone Z-score in obese boys but not in obese girls.

In obese subjects with %Fat ≥25, negative correlations of adiponectin with all parameters for obesity (BW, BMI, %Fat and FM) were demonstrated only in girls. These findings confirmed evidence reporting %Fat to be the better indicator of weight status to identify children and adolescents with unfavorable lipid profiles, mainly among girls (Oliosa et al., 2019), and a negative association between visceral fat and adiponectin in adolescents (Campos et al., 2018).

In addition, we also found moderately negative correlations of 25(OH)D with heel bone Z-score and SOS in obese girls despite serum 25(OH)D levels being in a normal range, which may be not consistent with previous studies reporting a high prevalence of vitamin D insufficiency in overweight and obese children and adolescents (Alamri et al., 2017); vitamin D insufficiency and excessive fat accumulation had mutually negative effects (Hyppönen and Power, 2006).

However, the limitations of this study included the small number of subjects and the short span of age; thus, the statistics might not be sufficient to refer to Thai obese adolescents along with comparisons between obese and nonobese subjects. Further research will be studied to confirm these relationships, focusing on negative associations between %Fat and biomarkers for bone and fat metabolism and a favorable effect of MM on calcaneus bone mass in obese girls that may associate with their earlier puberty, sex steroids and growth hormone.

5. Conclusion

The associations of body fat and bone parameters with OC, adiponectin, 25(OH)D, and insulin were sex-specific, with greater clarity when %Fat was used instead of BMI to classify obesity, and excessive body fat might be a risk factor of bone mass.

Ethics approval and consent to participate

The study protocol was approved by the Ethics Committee on Human Rights Related to Research Involving Human Subjects, Walailak University, Thailand (WU-IRG-62-018 and Protocol Number WU-EC-MD-2-048-62). A consent form was obtained from all subjects or their legal representative before enrollment.

Funding

This work was supported by the Research Institute for Health Sciences, Walailak University, Thailand under the contract no. WU-IRG-62-018.

CRediT authorship contribution statement

Rapheeporn Khwanchuea: Conceptualization; Formal analysis; Funding acquisition; Investigation; Methodology; Project administration; Validation; Writing - original draft. Chuchard Punsawad: Methodology; Resources; Funding acquisition; Writing - review & editing.

Declaration of competing interest

The results presented in this paper have not been published previously in whole or part. All the authors declared no competing interests.

Acknowledgments

Acknowledgement

We would like to sincerely thank all participants for their cooperation and wish to deeply thank the teachers and staff of secondary schools for their help and collaboration.

Contributor Information

Rapheeporn Khwanchuea, Email: krapheep@wu.ac.th, krapheep@mail.wu.ac.th.

Chuchard Punsawad, Email: chuchard.pu@wu.ac.th.

References

  1. Alamri Nejood Saied, Hussain A.H.J., Alzein Esra Hassan, Al Jareedan Talib Wasil, Ahmed Dawood, Mohammed A., Al Hamad Mona Ahmed, Alkhalifah S., Zakaria Nadia, Alastal ziz Mohammed, Asery Salha Ali Ahmad, Aldoobi Ghadeer Sameer, Al Faraj Zainab Mohammed, HussainMuzaffar Sukayna, Muzaffar Ibrahim Hussain, Al Momen Hajer Mohammed. Leptin as a potential biomarker for childhood obesity. EC Paediatr. 2017:435–446. [Google Scholar]
  2. Armaiz-Flores Sara A., Kelly Nichole R., Galescu Ovidiu A., Demidowich Andrew P., Altschul Anne M., Brady Sheila M., Hubbard Van S., Pickworth Courtney K., Tanofsky-Kraff Marian, Shomaker Lauren B., Reynolds James C., Yanovski Jack A. Evaluating weight status and sex as moderators of the Association of Serum Leptin with Bone Mineral Density in Children and Adolescents. Horm. Res. Paediatr. 2017;87(4):233–243. doi: 10.1159/000459623. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Baroncelli G.I. Quantitative ultrasound methods to assess bone mineral status in children: technical characteristics, performance, and clinical application. Pediatr. Res. 2008;63(3):220–228. doi: 10.1203/PDR.0b013e318163a286. [DOI] [PubMed] [Google Scholar]
  4. Campos R., Masquio D.C.L., Corgosinho F.C., Carvalho-Ferreira J.P., Molin Netto B.D., Clemente A.P.G., Tock L., Tufik S., Mello M.T., Dâmaso A.R. Relationship between adiponectin and leptin on osteocalcin in obese adolescents during weight loss therapy. Arch. Endocrinol. Metab. 2018;62(3):275–284. doi: 10.20945/2359-3997000000039. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Cao J.J. Effects of obesity on bone metabolism. J. Orthop. Surg. Res. 2011;6(1):30. doi: 10.1186/1749-799X-6-30. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Chin K.Y., Ima-Nirwana S. Calcaneal quantitative ultrasound as a determinant of bone health status: what properties of bone does it reflect? Int. J. Med. Sci. 2013;10(12):1778–1783. doi: 10.7150/ijms.6765. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Dimitri P. Fat and bone in children - where are we now? Ann. Pediatr. Endocrinol. Metab. 2018;23(2):62–69. doi: 10.6065/apem.2018.23.2.62. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Dimitri P., Bishop N., Walsh J.S., Eastell R. Obesity is a risk factor for fracture in children but is protective against fracture in adults: a paradox. Bone. 2012;50(2):457–466. doi: 10.1016/j.bone.2011.05.011. [DOI] [PubMed] [Google Scholar]
  9. Dobroch J., Cieśluk K., Sawicka-Żukowska M., Krawczuk-Rybak M. Body composition measurements in paediatrics - a review. Part 1. Pediatr. Endocrinol. Diabetes Metab. 2018;24(4):185–190. doi: 10.5114/pedm.2018.83365. [DOI] [PubMed] [Google Scholar]
  10. Faienza Maria Felicia, D’Amato Gabriele, Chiarito Mariangela, Colaianni Graziana, Colucci Silvia, Grano Maria, Corbo Filomena, Brunetti Giacomina. Mechanisms involved in childhood obesity-related bone fragility. Front. Endocrinol. 2019;10(269) doi: 10.3389/fendo.2019.00269. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Fernandes T.A.P., Gonçalves L.M.L., Brito J.A.A. Relationships between bone turnover and energy metabolism. J. Diabetes Res. 2017;2017:9021314. doi: 10.1155/2017/9021314. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Fintini Danilo, Cianfarani Stefano, Cofini Marta, Andreoletti Angela, Ubertini Grazia Maria, Cappa Marco, Manco Melania. The bones of children with obesity. Front. Endocrinol. 2020;11:200. doi: 10.3389/fendo.2020.00200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Freedman D.S., Wang J., Maynard L.M., Thornton J.C., Mei Z., Pierson R.N., Dietz W.H., Horlick M. Relation of BMI to fat and fat-free mass among children and adolescents. Int. J. Obes. 2005;29(1):1–8. doi: 10.1038/sj.ijo.0802735. [DOI] [PubMed] [Google Scholar]
  14. Giudici K.V. PhD, Fisberg R.M., Marchioni D.M.L., Peters B.S.E., Martini L.A. Crosstalk between bone and fat tissue: associations between vitamin D, osteocalcin, adipokines, and markers of glucose metabolism among adolescents. J. Am. Coll. Nutr. 2017;36(4):273–280. doi: 10.1080/07315724.2016.1274923. [DOI] [PubMed] [Google Scholar]
  15. Hyppönen E., Power C. Vitamin D status and glucose homeostasis in the 1958 British birth cohort: the role of obesity. Diabetes Care. 2006;29(10):2244–2246. doi: 10.2337/dc06-0946. [DOI] [PubMed] [Google Scholar]
  16. Jaruratanasirikul Somchit, Chanpong Atchariya, Tassanakijpanich Nuttaporn, Sriplung Hutcha. Declining age of puberty of school girls in southern Thailand. World J. Pediatr. 2014;10(3):256–261. doi: 10.1007/s12519-014-0472-2. [DOI] [PubMed] [Google Scholar]
  17. Júnior Ismael Forte Freitas, Cardoso Jefferson Rosa, Destro Christofaro Diego G., Codogno Jamile Sanches, de Moraes Augusto César Ferreira, Fernandes Rômulo Araújo. The relationship between visceral fat thickness and bone mineral density in sedentary obese children and adolescents. BMC Pediatr. 2013;13(1):1–5. doi: 10.1186/1471-2431-13-37. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Kouda Katsuyasu, Ohara Kumiko, Fujita Yuki, Nakamura Harunobu, Tachiki Takahiro, Iki Masayuki. Relationships between serum leptin levels and bone mineral parameters in school-aged children: a 3-year follow-up study. J. Bone Miner. Metab. 2019;37(1):152–160. doi: 10.1007/s00774-018-0909-0. [DOI] [PubMed] [Google Scholar]
  19. Lee Kayoung. Sex-specific relationships between insulin resistance and bone mineral content in Korean adolescents. J. Bone Miner. Metab. 2013;31(2):177–182. doi: 10.1007/s00774-012-0396-7. [DOI] [PubMed] [Google Scholar]
  20. Loomba-Albrecht L.A., Styne D.M. Effect of puberty on body composition. Curr. Opin. Endocrinol. Diabetes Obes. 2009;16(1):10–15. doi: 10.1097/med.0b013e328320d54c. [DOI] [PubMed] [Google Scholar]
  21. Maggio A.B., Belli D.C., Puigdefabregas J.W., Rizzoli R., Farpour-Lambert N.J., Beghetti M., McLin V.A. High bone density in adolescents with obesity is related to fat mass and serum leptin concentrations. J. Pediatr. Gastroenterol. Nutr. 2014;58(6):723–728. doi: 10.1097/mpg.0000000000000297. [DOI] [PubMed] [Google Scholar]
  22. Nakavachara P., Pooliam J., Weerakulwattana L., Kiattisakthavee P., Chaichanwattanakul K., Manorompatarasarn R., Chokephaibulkit K., Viprakasit V. A normal reference of bone mineral density (BMD) measured by dual energy X-ray absorptiometry in healthy thai children and adolescents aged 5-18 years: a new reference for southeast Asian populations. PLoS One. 2014;9(5) doi: 10.1371/journal.pone.0097218. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. O’Connor E.M., Edel D. Osteocalcin: the extra-skeletal role of a vitamin K-dependent protein in glucose metabolism. J. Nutr. Intermed. Metab. 2017;7:8–13. doi: 10.1016/j.jnim.2017.01.001. [DOI] [Google Scholar]
  24. Oliosa P.R., Zaniqueli D., Alvim R.O., Barbosa M.C.R., Mill J.G. Body fat percentage is better than indicators of weight status to identify children and adolescents with unfavorable lipid profile. J. Pediatr. 2019;95(1):112–118. doi: 10.1016/j.jped.2017.11.003. [DOI] [PubMed] [Google Scholar]
  25. de Onis M., Onyango A.W., Borghi E., Siyam A., Nishida C., Siekmann J. Development of a WHO growth reference for school-aged children and adolescents. Bull. World Health Organ. 2007;85(9):660–667. doi: 10.2471/blt.07.043497. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Park S.W., Nam G.E., Jung D.W., Yoon S.J., Han K., Park Y.G., Choi J.S., Lee J.E., Sang J.E., Yoon Y.J., Kim D.H. Association of lipid parameters and insulin resistance with bone health in South Korean adolescents. Osteoporos. Int. 2016;27(2):635–642. doi: 10.1007/s00198-015-3306-8. [DOI] [PubMed] [Google Scholar]
  27. Punthakee Z., Delvin E.E., O’Loughlin J., Paradis G., Levy E., Platt R.W., Lambert M. Adiponectin, adiposity, and insulin resistance in children and adolescents. J. Clin. Endocrinol. Metab. 2006;91(6):2119–2125. doi: 10.1210/jc.2005-2346. [DOI] [PubMed] [Google Scholar]
  28. Reinehr T., Roth C.L. A new link between skeleton, obesity and insulin resistance: relationships between osteocalcin, leptin and insulin resistance in obese children before and after weight loss. Int. J. Obes. 2010;34(5):852–858. doi: 10.1038/ijo.2009.282. [DOI] [PubMed] [Google Scholar]
  29. Rokoff Lisa B., Rifas-Shiman Sheryl L., Switkowski Karen M., Young Jessica G., Rosen Clifford J., Oken Emily, Fleisch Abby F. Body composition and bone mineral density in childhood. Bone. 2019;121:9–15. doi: 10.1016/j.bone.2018.12.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Schraders Katie, Zatta Giancarla, Kruger Marlena, Coad Jane, Weber Janet, Brough Louise, Thomson Jasmine. Quantitative ultrasound and dual X-ray absorptiometry as indicators of bone mineral density in young women and nutritional factors affecting it. Nutrients. 2019;11(10):2336. doi: 10.3390/nu11102336. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Talma H., Chinapaw M.J., Bakker B., HiraSing R.A., Terwee C.B., Altenburg T.M. Bioelectrical impedance analysis to estimate body composition in children and adolescents: a systematic review and evidence appraisal of validity, responsiveness, reliability and measurement error. Obes. Rev. 2013;14(11):895–905. doi: 10.1111/obr.12061. [DOI] [PubMed] [Google Scholar]
  32. World Health Organization. Regional Office for the Western, Pacific . Health Communications Australia; Sydney: 2000. The Asia-Pacific Perspective : Redefining Obesity and Its Treatment.https://apps.who.int/iris/handle/10665/206936 [Google Scholar]

Articles from Bone Reports are provided here courtesy of Elsevier

RESOURCES