J Bone Miner Metab DOI 10.1007/s00774-014-0615-5
ORIGINAL ARTICLE
Reference data and percentile curves of body composition measured with dual energy X-ray absorptiometry in healthy Chinese children and adolescents Bin Guo • Yi Xu • Jian Gong • Yongjin Tang Jingjie Shang • Hao Xu
•
Received: 21 May 2014 / Accepted: 26 June 2014 Ó The Japanese Society for Bone and Mineral Research and Springer Japan 2014
Abstract Measurements of body composition by dualenergy X-ray absorptiometry (DXA) have evident value in evaluating skeletal and muscular status in growing children and adolescents. This study aimed to generate age-related trends for body composition in Chinese children and adolescents, and to establish gender-specific reference percentile curves for the assessment of muscle-bone status. A total of 1541 Chinese children and adolescents aged from 5 to 19 years were recruited from southern China. Bone mineral content (BMC), lean mass (LM) and fat mass (FM) were measured for total body and total body less head (TBLH). After 14 years, total body LM was significantly higher in boys than girls (p \ 0.001). However, total body FM was significantly higher in girls than boys in age groups 13–19 years (p \ 0.01). Both LM and FM were consistent independent predictors of total body and subcranial bone mass in both sexes, even after adjustment for the well-known predictors of BMC. The results of multiple linear regression identified LM as the stronger predictor of total body and subcranial skeleton BMC while the fat mass contributed less. For all the subjects, significant positive correlations were observed between total body LM, height, total body BMC and subcranial BMC (p \ 0.01). Subcranial BMC had a better correlation with LM than total body BMC. We have also presented gender-specific percentile curves for LM-for-height and BMC-for-LM which could B. Guo J. Gong Y. Tang J. Shang H. Xu (&) Department of Nuclear Medicine, The First Affiliated Hospital, Jinan University, No.613, West Huangpu Road, Guangzhou 510630, China e-mail: [email protected] Y. Xu Department of Clinical Medicine, Medical College, Jinan University, Guangzhou 510630, China
be used to evaluate and follow various pediatric disorders with skeletal manifestations in this population. Keywords reference
Body composition DXA Children Normal
Introduction Dual-energy X-ray absorptiometry (DXA) is the most widely used method for diagnosing osteoporosis in adults. Due to its speed, high precision, accuracy, safety, low cost, low radiation exposure and widespread availability, DXA has become the gold standard for measuring bone mineral density (BMD) and bone mineral content (BMC) at regional sites and the total body in children and adolescents throughout the world [1]. However, the use of this method in childhood and adolescence is not without pitfalls due to bone growth in these periods [2]. The main problem in interpreting densitometry in children is the effect of bone size on BMC and BMD. In other words, it tends to underestimate bone density in small subjects and overestimate it in larger subjects [3]. Alternatively, for a more appropriate way of describing bone mass in children, one of the approaches to correct for bone size is the pediatric skeletal assessment tools adjusted for body size (ratios of height for age, BMC for bone area (BA), and BA for height, which may have some useful clinical applications [4]. We have previously reported DXA normative data for Chinese children and adolescents aged 5–19 years based on Molgaard et al.’s method [5]. Apart from quantifying areal BMD, DXA provides information for a three-compartment model of body composition: BMC, lean mass (LM), and fat mass (FM). This
123
J Bone Miner Metab
technique has clear advantages for measurement of LM (an estimate of muscle mass) [6–8]. Several studies have shown that there is a high correlation between BMC and LM in healthy children and adolescents [9–15], consistent with the functional muscle–bone unit theory [16]. Crabtree et al. [17] proposed a two-step algorithm to investigate the relationship between LM and BMC in children with chronic disease. First the amount of LM (a measure of muscle mass) relative to height is estimated and then the amount of BMC relative to LM is assessed. The new LM assessment can help to identify muscle weakness or muscle–bone imbalance. Only a few studies provide normative data of muscle–bone relationship indicators based on the new LM assessment using the DXA method [12–14]. Furthermore, the Z-score can be used to express precisely comparison with peers matched for body composition results based on the normal reference range. There are no published normal reference values on the new LM assessment tools in Chinese children and adolescents. Thus, an ethnicity-matched pediatric reference database needs to be established for use. Therefore, the main purposes of this study are as follows: (1) to determine age-specific and gender-specific normative reference values for body composition parameters (whole body LM and FM) in relation to age and height, and (2) to provide more effective LM assessment tools for DXA data interpretation which can evaluate and follow pediatric skeletal status in Chinese children and adolescents in a rational way.
Materials and methods Subjects In this cross-sectional study, we evaluated the muscle–bone relationship in 1541 healthy Chinese school children and adolescents aged 5–19 years (777 boys, 764 girls) recruited from four local schools in the Guangzhou district and one school in the Jiaxing district in southern China. All participating children were of Chinese ethnicity. Participants included in the study were between the 3rd and 97th percentiles for height and weight on current growth reference curves [18, 19]. The schools required the children to have outdoor exercise for about an hour each day; none of them had undertaken long-term intensive sports training. Subjects were excluded from the study if they had: (1) a history of metabolic disease or other medical disorders affecting bone growth and metabolism; (2) a history of use of medications affecting bone growth and metabolism; (3) a history of fracture; or (4) a body mass index (BMI) C30 kg/m2. Informed consent was obtained from all participants and their parents. This study was approved by the Ethics Committee of the First Affiliated Hospital, Jinan University.
123
Anthropometric and DXA measurements Anthropometric and DXA measurements were obtained for the children and adolescents during the same visit. Weight was measured using platform digital scales with a precision of 0.1 kg, and height was recorded with a stadiometer to the nearest 0.1 cm. Total body composition including BMC, LM and FM was measured with a Lunar Prodigy DXA bone densitometer (GE Healthcare, Madison, WI), and data were analyzed using enCORE software (ver. 10.0, standard-array mode). The total body less head (TBLH) variables were determined with the head region of interest removed from analysis. The precision for total body BMD, BMC, FM and LM was 0.50, 1.05, 1.95 and 0.93 % (expressed as the root-mean-square percent coefficient of variation), determined by duplicate scans with repositioning between each measurement in 30 volunteer subjects. A daily quality assurance scan was conducted by scanning an aluminum spine phantom according to the manufacturer’s instructions. All DXA measurements were performed by the well-trained technologist throughout the study. Statistical analysis Descriptive statistics were used to analyze baseline characteristics and measurements. Students t tests were conducted to evaluate differences in various parameters between male and female subjects. Pearson’s correlation coefficients (r) were calculated to assess the possible correlations among various parameters. The relationships of LM and FM to total body and TBLH bone mass were evaluated by multiple linear regression analysis. The LM-for-height and BMC-for-LM percentile curves (3rd, 25th, 50th, 75th, and 97th) were developed using the LMS method as described by Cole and Green [20]. The LMS method summarizes the changing distribution by three curves representing the median (M), coefficient of variation (S), and the skewness (L) expressed as a Box–Cox power. These three values were estimated, and the curves were calculated using the formula: Measurement percentile = M (1 ? LSZ)1/L. where Z is the Z-score corresponding to a given percentile. The percentile curves were constructed using the lmsChartMaker program (ver. 2.3; Medical Research Council, UK). All the tests were 2-tailed, and a p value of less than 0.05 was considered statistically significant.
Results The baseline characteristics of the participating subjects have been published previously [5]. Table 1 summarizes height, LM, FM and BMC of the participating subjects for
66 85
36
66
68
91
70
55
28
41
34
41
39
38
19
5 6
7
8
9
10
11
12
13
14
15
16
17
18
19
172.1 (6.1)c
46.9 (4.0)c
46.5 (4.3)
c
c
169.5 (4.2)
45.5 (5.3)c
168.4 (5.0)c
45.3 (4.8)
c
168.3 (5.4)
42.7 (4.9)c
c
41.2 (6.6)
c
166.4 (6.7)c
165.7 (8.1)
c
32.4 (6.1)
27.6 (4.7)
153.5 (9.1)
25.8 (3.8)
145.6 (8.2)a
23.7 (3.1)b
137.8 (7.6)a
143.0 (6.9)
21.7 (2.4)
20.5 (1.9)
2546.93 (327.44)c
2473.95 (418.15)
c
2338.91 (315.51)c
2307.24 (327.72)c
2032.87 (367.44)
41
55
72
38
31
43
26
41
61
71
51
62
41
63 68
Girls n
a = p \ 0.05; b = p \ 0.01; c = p \ 0.001. compared with girls of the same age group (unpaired-sample t tests)
9.6 (5.4)b
6.6 (4.5)
c
7.0 (3.0)c
6.2 (3.2)c
6.8 (3.3)c
1921.20 (437.81)
1415.66 (257.94)
5.6 (3.7)b 5.3 (2.8)c
1224.16 (244.02)a
1199.16 (215.10)
1061.12 (174.92)
932.71 (155.85)
869.55 (120.84)
770.48 (138.39)
631.05 (96.83) 698.41 (121.12)
BMC (g)
6.5 (4.7)
5.9 (4.0)
5.5 (3.6)
4.6 (4.3)
3.1 (1.9)
3.1 (2.0)
18.4 (2.1)c b
2.4 (1.5) 2.5 (1.6)a
15.6 (1.6)c 16.9 (2.1)c b
FM (kg)
LM (kg)
131.9 (5.2)
127.9 (0.6)
122.3 (6.3)
113.8 (4.9) 117.5 (5.5)
Height (cm)
LM lean mass, FM fat mass, BMC bone mineral content
Boys n
Age (years)
Table 1 Means and SDs of height, LM, FM, and BMC in children according to age groups
159.5 (7.2)
157.5 (5.3)
157.1 (6.2)
157.1 (5.8)
159.4 (5.8)
158.7 (6.0)
155.8 (7.5)
149.5 (6.2)
145.7 (7.6)
138.6 (6.9)
131.4 (6.0)
128.5 (6.6)
121.1 (7.5)
112.9 (5.6) 117.1 (4.5)
Height (cm)
32.9 (3.8)
32.2 (4.9)
32.0 (3.5)
32.1 (3.4)
32.5 (4.2)
32.7 (4.3)
31.4 (3.9)
27.6 (3.5)
25.5 (3.6)
22.2 (3.0)
19.8 (2.3)
19.2 (2.2)
16.7 (1.6)
14.3 (1.7) 15.5 (1.6)
LM (kg)
14.2 (5.5)
12.9 (3.4)
13.4 (3.8)
14.0 (4.5)
13.2 (5.4)
11.4 (4.7)
8.7 (3.9)
7.3 (3.4)
6.6 (3.2)
4.9 (2.1)
4.4 (1.8)
3.3 (1.6)
3.3 (1.7)
2.5 (1.1) 3.1 (1.6)
FM (kg)
2042.86 (354.43)
2010.35 (320.38)
1970.83 (299.99)
1964.39 (295.05)
1943.84 (312.03)
1862.99 (354.58)
1538.22 (282.12)
1333.02 (250.38)
1215.02 (219.09)
1013.65 (179.80)
909.36 (157.38)
862.83 (160.70)
745.02 (122.66)
599.45 (86.92) 664.06 (107.28)
BMC (g)
J Bone Miner Metab
123
J Bone Miner Metab
each gender and age group. For total body LM, significant gender differences were found in age groups 5–7 years (p \ 0.001), 8 years (p \ 0.01), 10 years (p \ 0.01) and 14–19 years (p \ 0.001). After 14 years, total body LM was significantly higher in boys than girls. However, for total body FM, the gender differences in these age groups were also observed in age groups 6 years (p \ 0.05), 8 years (p \ 0.01) 13 years (p \ 0.01), 14–18 years (p \ 0.001) and 19 years (p \ 0.01). After 13 years, total body FM was significantly higher in girls than boys. The relationship between body composition and bone mass was studied using the Pearson correlation coefficient and linear regression analysis. For all the subjects, significant positive correlations were observed between LM, FM and total body BMD, total body BMC, subcranial BMD, and subcranial BMC with r ranging from 0.451 to 0.977 (p \ 0.01). Bone mass for total body and subcranial skeleton as measured by BMC (g), BMD (g/cm2) and BMD/ height (g/cm3) was modeled as the dependent variable,
while age, height, LM and FM were treated as the independent variables. The model with the highest r2 value was chosen. Both LM and FM were consistent independent predictors of total body and subcranial bone mass in both sexes. The results of multiple linear regression identified LM as the stronger predictor of total body BMC and subcranial skeleton BMC while the FM contributed less as shown in Tables 2 and 3. For all the subjects, significant positive correlations were observed between total body LM, height, total body BMC and subcranial BMC (p \ 0.01). LM–height tended to be higher in boys than in girls. After evaluation of the different models, the exponential model was found to be the best fit for the association between LM and height and the linear model for total body and subcranial BMC and LM in both genders. The corresponding coefficients of determination (R2) were also calculated. Correlations between total body LM and height, and total body and subcranial BMC and LM are shown in Figs. 1 and 2,
Table 2 Multiple regression of total body and subcranial BMC, BMD, and BMD/height against age, height, LM and FM among boys Subcranial skeletona
Total body BMC
BMD
BMD/H
BMC
BMD
BMD/H
b
p
b
p
b
p
b
p
b
p
b
p
Age (years)
0.086
0.038
0.240
c
0.311
0.004
–
–
0.132
c
0.297
0.004
Height (cm)
0.494
c
0.239
c
-1.308
c
0.557
c
0.481
c
-0.537
c
LM (kg)
0.359
c
0.384
c
0.696
c
0.388
c
0.312
c
0.677
c
FM (kg)
0.064
c
0.062
0.002
0.088
0.012
0.060
c
0.071
c
0.172
c
model R2
0.892
0.753
0.279
0.901
0.873
0.343
BMC bone mineral content, BMD bone mineral density, BMD/H bone mineral density/height Total body with the head region of interest removed from analysis
a
c = p \ 0.001
Table 3 Multiple regression of total body and subcranial BMC, BMD, and BMD/height against age, height, LM and FM among girls Subcranial skeletona
Total body BMC
BMD
BMD/H
BMC
BMD
BMD/H
b
p
b
p
b
p
b
p
b
p
b
p
Age (years)
0.237
c
0.429
c
0.700
c
0.166
c
0.240
c
0.473
c
Height (cm) LM (kg)
–
–
-0.222
c
-1.901
c
0.061
0.032
–
–
-1.333
c
0.528
c
0.515
c
0.887
c
0.544
c
0.572
c
1.088
c
FM (kg)
0.259
c
0.220
c
0.370
c
0.257
c
0.178
c
0.355
c
model R2
0.932
0.796
0.415
0.843
BMC bone mineral content, BMD bone mineral density, BMD/H bone mineral density/height, a
Total body with the head region of interest removed from analysis
c = p \ 0.001
123
0.882
0.499
J Bone Miner Metab
Fig. 1 Correlations of total body lean mass calculated by DXA with height in boys (a) and girls (b)
Fig. 2 Correlations of total body and subcranial bone mineral content with lean mass calculated by DXA in boys (a) and girls (b)
respectively, for each gender group. The gender-specific LM-for-height percentile curves are displayed in Fig. 3. In general, the percentile curves for the two genders were
similar in shape. The LM-dependent percentile curves for BMC (Fig. 4) showed that BMC was closely associated with LM for total body and TBLH.
123
J Bone Miner Metab
Fig. 3 Lean mass percentile curves adjusted for height. Solid lines from the top represent the 97th, 75th, 50th, 25th, and 3rd percentiles
Discussion In this paper, we have produced age- and gender-specific normative body composition reference data for Chinese children and adolescents measured using a GE Lunar Prodigy DXA scanner. Furthermore, we have provided reference curves for LM–height and BMC–LM which will allow assessment of the skeletal and muscular status of Chinese children and adolescents using Crabtree’s approaches [17]. Ethnicity differences in body composition of children and adolescents has already been reported [21–23], proving that it is necessary to establish ethnicity-specific reference databases for the interpretation of pediatric DXA results. For example, Ellis et al. [22, 23] reported that LM was higher in black than in white boys and girls; no difference in LM was evident between white and Hispanic groups. Higher FM and body fat percentage values were observed in Hispanic boys and girls, even after correcting for body size. An increase in total body LM and FM with age is seen in our study, similar to that found in other studies [24–26]. Our data show that total body LM becomes significantly higher in boys than in girls from the age of 14 years, confirming previous results [26] and FM values become higher in girls than in boys from the age of 13 years, also in agreement with other studies [27]. In our literature review, puberty had a significant effect on body composition in boys and girls. For boys, greater LM may be caused by increased growth hormone and androgen concentrations. Synergism between growth hormone and androgens has already been described [28]. However, greater FM was discovered in girls after 13. It is generally accepted that body fat is a significant factor for female growth. Adipose
123
tissue may directly affect the menstrual cycle and ovulation, and hence fertility. Frisch and Revelle [29] suggested that a body fat of 17 % is needed to trigger the change in metabolic rate that leads to menarche, and a body fat of 22 % is required to maintain female reproductive ability [30]. In this study, participants were between the 3rd and 97th percentilea for height and weight on current standardized growth charts, roughly consistent with the mean values for Chinese children and adolescents. Therefore, we believe the data presented can represent LM and FM reference values and reflect the growth and development trends of a healthy population. In the present investigation, we adopted the method of Khosla et al. [31] to analyze the relationship between body composition and bone mass that is critically dependent on the specific parameter (i.e., BMC, BMD, BMD/height) in the analysis. Thus, using these correction factors tends to reduce the apparent influence of LM on BM. A number of body composition studies [14, 15, 17, 32–36] in children and adolescents have attempted to disentangle the independent relationships between LM and BM; their results are consistent with our own. Hogler W et al. [14] found that LM explained 96.8 % of the variation in total body BMC in boys and 95.1 % in girls. Pietrobelli et al. [32] showed that a 1 kg increase in LM was associated with a 0.04 kg increase in total body BMC on average, whereas a 1 kg increase in FM was associated with only a 0.009 kg increase in total body BMC. Arabi et al. [15], in a crosssectional study conducted with 363 school children aged 10–17 years, demonstrated that LM and FM are predictors of bone mass in boys and girls. The contribution of LM to BMC variance was larger than that of FM. Our results support the conclusion that both LM and FM are consistent independent predictors of total body and subcranial bone
J Bone Miner Metab
Fig. 4 Bone mineral content (BMC) percentile curves adjusted for lean mass. Solid lines from the top represent the 97th, 75th, 50th, 25th, and 3rd percentiles
mass in both sexes, even after adjustment for the wellknown predictors of BMC in healthy Chinese children and adolescents. Total body and subcranial BMC are more strongly associated with LM than with FM. Our data is in line with the notion that muscle development is correlated with skeletal development. All this suggests that analysis of bone parameters should be performed when considering muscle function. This study reports on the relationship between LM and height in subjects aged 5–19 years. Because muscle force is largely determined by body height [37, 38], muscle parameters should be measured in relationto body height during growth. Furthermore, it was found that LM values tend to plateau with age [22, 23, 26] but increase with increasing body height [12, 14, 17]. Our study reveals that LM values are positively and strongly related to height and
LM–height tends to be higher in boys than in girls, consistent with data published elsewhere [12, 14, 17]. In general, the LM assessment tools using DXA should be adjusted for height-. The International Society of Clinical Densitometry (ISCD) recommends that total body and TBLH BMD results should be adjusted for body size, using, for example, height, age or height-specific Z scores [2]. For this reason, height-dependent reference data for LM of the total body should be established for each gender group. Our study results indicate that total body and subcranial BMC and LM are highly correlated for each gender group, respectively, which is in line with the results from previous studies [12–14, 17, 39]. These statistical associations between LM and BMC both for total body and TBLH support the Frost mechanostat thesis that since muscle
123
J Bone Miner Metab
action delivers the largest loads and bone strains, reflecting the functional adaptation of bone to its function, then the increase in muscle development must precede and should determine the increase in BMC [39, 40]. Schoenau and Frost [41, 42] suggested that bone and muscle form a useful functional biological ‘‘muscle–bone unit’’ and that muscle– bone interactions during growth are key determinants of skeletal adaptation towards changing loads, which underlines the importance of evaluating bone and muscle tissues in the pathophysiology of childhood skeletal disorders. For example, LM assessment has been employed in studies of the muscular and skeletal status of children with idiopathic juvenile osteoporosis and osteogenesis imperfecta [43, 44], as well as by Hogler et al. in children with anorexia nervosa and growth hormone deficiency [14] and by Khadilkar in a female child presenting with stunting [13]. To evaluate skeletal status in children, the most accurate and reproducible skeletal sites for performing BMC and areal BMD measurements in this population are posteroanterior (PA) spine and TBLH as recommended by the ISCD [45]. Our study shows that subcranial BMC has a better correlation with LM than total body BMC and that subcranial skeleton parameters may be important indicators of muscular and skeletal status in children and adolescents. Pediatric body composition data are often difficult to interpret in children and adolescents. In a previous publication, a number of novel approaches have been proposed. Regarding application of the muscle–bone relationship, Scho¨nau et al. [39] suggested that measures of bone status (BMC) should be combined with those on muscle status (muscle cross-sectional area), using peripheral quantitative computed tomography (pQCT). The original idea was then extended by Ho¨gler et al. [14] to DXA which can differentiate of the origin of a low BMD or BMC/age, for example, short stature and primary, secondary, and mixed bone defects. With this in mind, Crabtree et al. has suggested another method for normalizing bone data [17], which shows how the associations between LM and height in addition to LM and BMC can be used to identify whether the primary abnormality is in muscle or bone. Data is analyzed in two steps: (1) the amount of LM relative to height is assessed (this reflects sacropenia); and (2) the amount of BMC relative to LM is assessed (reflecting the degree of osteopenia). In this study, we have adopted Crabtree et al.’s method [17] to develop percentile curves in addition to normative body composition data. The main perspective of our study is to provide new LM assessment tools helpful for elucidating the origin of low bone mass in children and adolescents, focusing on the functional biological ‘‘muscle–bone unit’’. We acknowledge some potential limitations to this study. The primary limitation is that it was a cross-sectional study and further longitudinal data need to be
123
obtained. Secondly, there are other limitations concerning DXA. Due to differences in scan modes, software version [46], and the calibration methods adopted by different DXA manufacturers [47, 48], variations between different DXA scanners are known to result in substantial differences in body composition measurements. Our reference data is limited only to results derived from the Lunar Prodigy DXA densitometer; these reference curves cannot be used with the results from other brands of DXA scanner. The final limitation is that we could not acquire the Tanner stage of the subjects. Furthermore, caution should be advised when attempting to utilize the provided normative values due to the differences in genetic background, nutritional habits, physical activity level, and other lifestyle choices of the study population. In summary, we have presented an ethnicity- and gender-specific body composition normal reference database as well as percentile curves for normalization of muscle– bone data by DXA for Chinese children and adolescents aged 5–19 years. To the best of our knowledge, these are the first total body normal reference data for Chinese children and adolescents. Therefore, it is worthwhile establishing the new LM assessment reference values allowing precise comparison with the given results and then calculating the Z score in Chinese children and adolescents. The results of this study may be used for clinical differentiation and assessment of children with different aetiologies of pediatric bone disorders. Further work will be required to evaluate this relationship with other clinical conditions in China as measured by the Lunar Prodigy DXA densitometer. Acknowledgments The authors would like to express their gratitude to all participating children and their parents. We are grateful to Dr. Qi Zhou, GE Healthcare Shanghai and Dr. Jing Xiang, First Hospital of Jiaxing for their useful comments and suggestions. We also thank the staff members of the Department of Nuclear Medicine, First Affiliated Hospital of Jinan University for excellent technical support. Conflict of interest None of the authors have any personal or financial conflicts of interest.
References 1. Bachrach LK (2005) Osteoporosis and measurement of bone mass in children and adolescents. Endocrinol Metab Clin North Am 34:521–535 2. Gordon CM, Bachrach LK, Carpenter TO, Crabtree N, El-Hajj FG, Kutilek S, Lorenc RS, Tosi LL, Ward KA, Ward LM, Kalkwarf HJ (2008) Dual energy X-ray absorptiometry interpretation and reporting in children and adolescents: The 2007 ISCD Pediatric Official Positions. J Clin Densitom 11:43–58 3. Fewtrell MS (2003) Bone densitometry in children assessed by dual X-ray absorptiometry: uses and pitfalls. Arch Dis Child 88:795–798
J Bone Miner Metab 4. Molgaard C, Thomsen BL, Prentice A, Cole TJ, Michaelsen KF (1997) Whole body bone mineral content in healthy children and adolescents. Arch Dis Child 76:9–15 5. Guo B, Xu Y, Gong J, Tang Y, Xu H (2013) Age trends of bone mineral density and percentile curves in healthy Chinese children and adolescents. J Bone Miner Metab 31:304–314 6. Binkovitz LA, Henwood MJ, Sparke P (2008) Pediatric DXA: technique, interpretation and clinical applications. Pediatr Radiol 38:S227–S239 7. Binkley TL, Berry R, Specker BL (2008) Methods for measurement of pediatric bone. Rev Endocr Metab Disord 9:95–106 8. Kim J, Wang Z, Heymsfield SB, Baumgartner RN, Gallagher D (2002) Total-body skeletal muscle mass: estimation by a new dual-energy X-ray absorptiometry method. Am J Clin Nutr 76:378–383 9. Arabi A, Nabulsi M, Maalouf J, Choucair M, Khalife H, Vieth R, El-Hajj Fuleihan G (2004) Bone mineral density by age, gender, pubertal stages, and socioeconomic status in healthy Lebanese children and adolescents. Bone 35:1169–1179 10. Hogler W, Briody J, Woodhead HJ, Chan A, Cowell CT (2003) Importance of lean mass in the interpretation of total body densitometry in children and adolescents. J Pediatr 143:81–88 11. Arabi A, Tamim H, Nabulsi M, Maalouf J, Khalife H, Choucair M, Vieth R, El-Hajj Fuleihan G (2004) Sex differences in the effect of body-composition variables on bone mass in healthy children and adolescents. Am J Clin Nutr 80:1428–1435 12. Ogle GD, Allen JR, Humphries IR, Lu PW, Briody JN, Morley K, Howman-Giles R, Cowell CT (1995) Body-composition assessment by dual-energy X-ray absorptiometry in subjects aged 4–26 years. Am J Clin Nutr 61:746–753 13. Ferretti JL, Capozza RF, Cointry GR, Garcia SL, Plotkin H, Alvarez FML, Zanchetta JR (1998) Gender-related differences in the relationship between densitometric values of whole-body bone mineral content and lean body mass in humans between 2 and 87 years of age. Bone 22:683–690 14. Pludowski P, Matusik H, Olszaniecka M, Lebiedowski M, Lorenc RS (2005) Reference values for the indicators of skeletal and muscular status of healthy Polish children. J Clin Densitom 8:164–177 15. Khadilkar AV, Sanwalka NJ, Chiplonkar SA, Khadilkar VV, Mughal MZ (2011) Normative data and percentile curves for dual energy X-ray absorptiometry in healthy Indian girls and boys aged 5-17 years. Bone 48:810–819 16. Frost HM, Schonau E (2000) The ‘‘muscle-bone unit’’ in children and adolescents: a 2000 overview. J Pediatr Endocrinol Metab 13:571–590 17. Crabtree NJ, Kibirige MS, Fordham JN, Banks LM, Muntoni F, Chinn D, Boivin CM, Shaw NJ (2004) The relationship between lean body mass and bone mineral content in paediatric health and disease. Bone 35:965–972 18. Li H, Ji CY, Zong XN, Zhang YQ (2009) Height and weight standardized growth charts for Chinese children and adolescents aged 0 to 18 years. Zhonghua Er Ke Za Zhi 47:487–492 19. Ministry of Education of the People’s Republic of China (2007) Report on the physical fitness and health surveillance of Chinese school students. Higher Education Press, Beijing 20. Cole TJ, Green PJ (1992) Smoothing reference centile curves: the LMS method and penalized likelihood. Stat Med 11:1305–1319 21. Nelson DA, Barondess DA (1997) Whole body bone, fat and lean mass in children: comparison of three ethnic groups. Am J Phys Anthropol 103:157–162 22. Ellis KJ (1997) Body composition of a young, multiethnic, male population. Am J Clin Nutr 66:1323–1331 23. Ellis KJ, Abrams SA, Wong WW (1997) Body composition of a young, multiethnic female population. Am J Clin Nutr 65:724–731
24. Sala A, Webber CE, Morrison J, Beaumont LF, Barr RD (2007) Whole-body bone mineral content, lean body mass, and fat mass measured by dual-energy X-ray absorptiometry in a population of normal Canadian children and adolescents. Can Assoc Radiol J 58:46–52 25. Alwis G, Rosengren B, Stenevi-Lundgren S, Duppe H, Sernbo I, Karlsson MK (2010) Normative dual energy X-ray absorptiometry data in Swedish children and adolescents. Acta Paediatr 99:1091–1099 26. Boot AM, Bouquet J, de Ridder MA, Krenning EP, de Muinck Keizer-Schrama SM (1997) Determinants of body composition measured by dual-energy X-ray absorptiometry in Dutch children and adolescents. Am J Clin Nutr 66:232–238 27. Kim K, Yun SH, Jang MJ, Oh KW (2013) Body fat percentile curves for Korean children and adolescents: a data from the Korea national health and nutrition examination survey 2009-2010. J Korean Med Sci 28:443–449 28. Martin LG, Grossman MS, Connor TB, Levitsky LL, Clark JW, Camitta FD (1979) Effect of androgen on growth hormone secretion and growth in boys with short stature. Acta Endocrinol 91:201–212 29. Frisch RE, Revelle R (1970) Height and weight at menarche and a hypothesis of critical body weights and adolescent events. Science 169:397–399 30. Frisch RE (1987) Body fat, menarche, fitness and fertility. Hum Reprod 2:521–533 31. Khosla S, Atkinson EJ, Riggs BL, Melton LJ 3rd (1996) Relationship between body composition and bone mass in women. J Bone Miner Res 11:857–863 32. Pietrobelli A, Faith MS, Wang J, Brambilla P, Chiumello G, Heymsfield SB (2002) Association of lean tissue and fat mass with bone mineral content in children and adolescents. Obes Res 10:56–60 33. Young D, Hopper JL, Nowson CA, Green RM, Sherwin AJ, Kaymakci B, Smid M, Guest CS, Larkins RG, Wark JD (1995) Determinants of bone mass in 10- to 26-year-old females: a twin study. J Bone Miner Res 10:558–567 34. Young D, Hopper JL, Macinnis RJ, Nowson CA, Hoang NH, Wark JD (2001) Changes in body composition as determinants of longitudinal changes in bone mineral measures in 8 to 26-yearold female twins. Osteoporos Int 12:506–515 35. Ilich JZ, Skugor M, Hangartner T, Baoshe A, Matkovic V (1998) Relation of nutrition, body composition and physical activity to skeletal development: a cross-sectional study in preadolescent females. J Am Coll Nutr 17:136–147 36. Valdimarsson O, Kristinsson JO, Stefansson SO, Valdimarsson S, Sigurdsson G (1999) Lean mass and physical activity as predictors of bone mineral density in 16-20-year old women. J Intern Med 245:489–496 37. Round JM, Jones DA, Honour JW, Nevill AM (1999) Hormonal factors in the development of differences in strength between boys and girls during adolescence: a longitudinal study. Ann Hum Biol 26:49–62 38. Parker DF, Round JM, Sacco P, Jones DA (1990) A cross-sectional survey of upper and lower limb strength in boys and girls during childhood and adolescence. Ann Hum Biol 17:199–211 39. Schoenau E, Neu CM, Beck B, Manz F, Rauch F (2002) Bone mineral content per muscle cross-sectional area as an index of the functional muscle-bone unit. J Bone Miner Res 17:1095–1101 40. Schoenau E (2005) From mechanostat theory to development of the ‘‘Functional Muscle-Bone-Unit’’. J Musculoskelet Neuronal Interact 5:232–238 41. Schoenau E, Frost HM (2002) The ‘‘muscle-bone unit’’ in children and adolescents. Calcif Tissue Int 70:405–407 42. Fricke O, Schoenau E (2007) The ‘Functional Muscle-Bone Unit’: probing the relevance of mechanical signals for bone
123
J Bone Miner Metab development in children and adolescents. Growth Horm IGF Res 17:1–9 43. Pludowski P, Lebiedowski M, Olszaniecka M, Marowska J, Matusik H, Lorenc RS (2006) Idiopathic juvenile osteoporosis–– an analysis of the muscle-bone relationship. Osteoporos Int 17:1681–1690 44. Pludowski P, Matusik HLRS (2003) Evaluation of musculoskeletal system in OI and IJO subjects. Osteoporosis Int 14:S31– S32 45. Khan AA, Bachrach L, Brown JP, Hanley DA, Josse RG, Kendler DL, Leib ES, Lentle BC, Leslie WD, Lewiecki EM, Miller PD, Nicholson RL, O’Brien C, Olszynski WP, Theriault MY, Watts NB (2004) Canadian panel of the international society of clinical
123
densitometry standards and guidelines for performing central dual-energy X-ray absorptiometry in premenopausal women, men, and children. J Clin Densitom 7:51–64 46. Laskey MA, Prentice A (1999) Comparison of adult and paediatric spine and whole body software for the lunar dual energy X-ray absorptiometer. Br J Radiol 72:967–976 47. Tothill P, Hannan WJ (2002) Bone mineral and soft tissue measurements by dual-energy X-ray absorptiometry during growth. Bone 31:492–496 48. Tothill P, Avenell A, Love J, Reid DM (1994) Comparisons between hologic, lunar and norland dual-energy X-ray absorptiometers and other techniques used for whole-body soft tissue measurements. Eur J Clin Nutr 48:781–794
ORIGINAL ARTICLE
Reference data and percentile curves of body composition measured with dual energy X-ray absorptiometry in healthy Chinese children and adolescents Bin Guo • Yi Xu • Jian Gong • Yongjin Tang Jingjie Shang • Hao Xu
•
Received: 21 May 2014 / Accepted: 26 June 2014 Ó The Japanese Society for Bone and Mineral Research and Springer Japan 2014
Abstract Measurements of body composition by dualenergy X-ray absorptiometry (DXA) have evident value in evaluating skeletal and muscular status in growing children and adolescents. This study aimed to generate age-related trends for body composition in Chinese children and adolescents, and to establish gender-specific reference percentile curves for the assessment of muscle-bone status. A total of 1541 Chinese children and adolescents aged from 5 to 19 years were recruited from southern China. Bone mineral content (BMC), lean mass (LM) and fat mass (FM) were measured for total body and total body less head (TBLH). After 14 years, total body LM was significantly higher in boys than girls (p \ 0.001). However, total body FM was significantly higher in girls than boys in age groups 13–19 years (p \ 0.01). Both LM and FM were consistent independent predictors of total body and subcranial bone mass in both sexes, even after adjustment for the well-known predictors of BMC. The results of multiple linear regression identified LM as the stronger predictor of total body and subcranial skeleton BMC while the fat mass contributed less. For all the subjects, significant positive correlations were observed between total body LM, height, total body BMC and subcranial BMC (p \ 0.01). Subcranial BMC had a better correlation with LM than total body BMC. We have also presented gender-specific percentile curves for LM-for-height and BMC-for-LM which could B. Guo J. Gong Y. Tang J. Shang H. Xu (&) Department of Nuclear Medicine, The First Affiliated Hospital, Jinan University, No.613, West Huangpu Road, Guangzhou 510630, China e-mail: [email protected] Y. Xu Department of Clinical Medicine, Medical College, Jinan University, Guangzhou 510630, China
be used to evaluate and follow various pediatric disorders with skeletal manifestations in this population. Keywords reference
Body composition DXA Children Normal
Introduction Dual-energy X-ray absorptiometry (DXA) is the most widely used method for diagnosing osteoporosis in adults. Due to its speed, high precision, accuracy, safety, low cost, low radiation exposure and widespread availability, DXA has become the gold standard for measuring bone mineral density (BMD) and bone mineral content (BMC) at regional sites and the total body in children and adolescents throughout the world [1]. However, the use of this method in childhood and adolescence is not without pitfalls due to bone growth in these periods [2]. The main problem in interpreting densitometry in children is the effect of bone size on BMC and BMD. In other words, it tends to underestimate bone density in small subjects and overestimate it in larger subjects [3]. Alternatively, for a more appropriate way of describing bone mass in children, one of the approaches to correct for bone size is the pediatric skeletal assessment tools adjusted for body size (ratios of height for age, BMC for bone area (BA), and BA for height, which may have some useful clinical applications [4]. We have previously reported DXA normative data for Chinese children and adolescents aged 5–19 years based on Molgaard et al.’s method [5]. Apart from quantifying areal BMD, DXA provides information for a three-compartment model of body composition: BMC, lean mass (LM), and fat mass (FM). This
123
J Bone Miner Metab
technique has clear advantages for measurement of LM (an estimate of muscle mass) [6–8]. Several studies have shown that there is a high correlation between BMC and LM in healthy children and adolescents [9–15], consistent with the functional muscle–bone unit theory [16]. Crabtree et al. [17] proposed a two-step algorithm to investigate the relationship between LM and BMC in children with chronic disease. First the amount of LM (a measure of muscle mass) relative to height is estimated and then the amount of BMC relative to LM is assessed. The new LM assessment can help to identify muscle weakness or muscle–bone imbalance. Only a few studies provide normative data of muscle–bone relationship indicators based on the new LM assessment using the DXA method [12–14]. Furthermore, the Z-score can be used to express precisely comparison with peers matched for body composition results based on the normal reference range. There are no published normal reference values on the new LM assessment tools in Chinese children and adolescents. Thus, an ethnicity-matched pediatric reference database needs to be established for use. Therefore, the main purposes of this study are as follows: (1) to determine age-specific and gender-specific normative reference values for body composition parameters (whole body LM and FM) in relation to age and height, and (2) to provide more effective LM assessment tools for DXA data interpretation which can evaluate and follow pediatric skeletal status in Chinese children and adolescents in a rational way.
Materials and methods Subjects In this cross-sectional study, we evaluated the muscle–bone relationship in 1541 healthy Chinese school children and adolescents aged 5–19 years (777 boys, 764 girls) recruited from four local schools in the Guangzhou district and one school in the Jiaxing district in southern China. All participating children were of Chinese ethnicity. Participants included in the study were between the 3rd and 97th percentiles for height and weight on current growth reference curves [18, 19]. The schools required the children to have outdoor exercise for about an hour each day; none of them had undertaken long-term intensive sports training. Subjects were excluded from the study if they had: (1) a history of metabolic disease or other medical disorders affecting bone growth and metabolism; (2) a history of use of medications affecting bone growth and metabolism; (3) a history of fracture; or (4) a body mass index (BMI) C30 kg/m2. Informed consent was obtained from all participants and their parents. This study was approved by the Ethics Committee of the First Affiliated Hospital, Jinan University.
123
Anthropometric and DXA measurements Anthropometric and DXA measurements were obtained for the children and adolescents during the same visit. Weight was measured using platform digital scales with a precision of 0.1 kg, and height was recorded with a stadiometer to the nearest 0.1 cm. Total body composition including BMC, LM and FM was measured with a Lunar Prodigy DXA bone densitometer (GE Healthcare, Madison, WI), and data were analyzed using enCORE software (ver. 10.0, standard-array mode). The total body less head (TBLH) variables were determined with the head region of interest removed from analysis. The precision for total body BMD, BMC, FM and LM was 0.50, 1.05, 1.95 and 0.93 % (expressed as the root-mean-square percent coefficient of variation), determined by duplicate scans with repositioning between each measurement in 30 volunteer subjects. A daily quality assurance scan was conducted by scanning an aluminum spine phantom according to the manufacturer’s instructions. All DXA measurements were performed by the well-trained technologist throughout the study. Statistical analysis Descriptive statistics were used to analyze baseline characteristics and measurements. Students t tests were conducted to evaluate differences in various parameters between male and female subjects. Pearson’s correlation coefficients (r) were calculated to assess the possible correlations among various parameters. The relationships of LM and FM to total body and TBLH bone mass were evaluated by multiple linear regression analysis. The LM-for-height and BMC-for-LM percentile curves (3rd, 25th, 50th, 75th, and 97th) were developed using the LMS method as described by Cole and Green [20]. The LMS method summarizes the changing distribution by three curves representing the median (M), coefficient of variation (S), and the skewness (L) expressed as a Box–Cox power. These three values were estimated, and the curves were calculated using the formula: Measurement percentile = M (1 ? LSZ)1/L. where Z is the Z-score corresponding to a given percentile. The percentile curves were constructed using the lmsChartMaker program (ver. 2.3; Medical Research Council, UK). All the tests were 2-tailed, and a p value of less than 0.05 was considered statistically significant.
Results The baseline characteristics of the participating subjects have been published previously [5]. Table 1 summarizes height, LM, FM and BMC of the participating subjects for
66 85
36
66
68
91
70
55
28
41
34
41
39
38
19
5 6
7
8
9
10
11
12
13
14
15
16
17
18
19
172.1 (6.1)c
46.9 (4.0)c
46.5 (4.3)
c
c
169.5 (4.2)
45.5 (5.3)c
168.4 (5.0)c
45.3 (4.8)
c
168.3 (5.4)
42.7 (4.9)c
c
41.2 (6.6)
c
166.4 (6.7)c
165.7 (8.1)
c
32.4 (6.1)
27.6 (4.7)
153.5 (9.1)
25.8 (3.8)
145.6 (8.2)a
23.7 (3.1)b
137.8 (7.6)a
143.0 (6.9)
21.7 (2.4)
20.5 (1.9)
2546.93 (327.44)c
2473.95 (418.15)
c
2338.91 (315.51)c
2307.24 (327.72)c
2032.87 (367.44)
41
55
72
38
31
43
26
41
61
71
51
62
41
63 68
Girls n
a = p \ 0.05; b = p \ 0.01; c = p \ 0.001. compared with girls of the same age group (unpaired-sample t tests)
9.6 (5.4)b
6.6 (4.5)
c
7.0 (3.0)c
6.2 (3.2)c
6.8 (3.3)c
1921.20 (437.81)
1415.66 (257.94)
5.6 (3.7)b 5.3 (2.8)c
1224.16 (244.02)a
1199.16 (215.10)
1061.12 (174.92)
932.71 (155.85)
869.55 (120.84)
770.48 (138.39)
631.05 (96.83) 698.41 (121.12)
BMC (g)
6.5 (4.7)
5.9 (4.0)
5.5 (3.6)
4.6 (4.3)
3.1 (1.9)
3.1 (2.0)
18.4 (2.1)c b
2.4 (1.5) 2.5 (1.6)a
15.6 (1.6)c 16.9 (2.1)c b
FM (kg)
LM (kg)
131.9 (5.2)
127.9 (0.6)
122.3 (6.3)
113.8 (4.9) 117.5 (5.5)
Height (cm)
LM lean mass, FM fat mass, BMC bone mineral content
Boys n
Age (years)
Table 1 Means and SDs of height, LM, FM, and BMC in children according to age groups
159.5 (7.2)
157.5 (5.3)
157.1 (6.2)
157.1 (5.8)
159.4 (5.8)
158.7 (6.0)
155.8 (7.5)
149.5 (6.2)
145.7 (7.6)
138.6 (6.9)
131.4 (6.0)
128.5 (6.6)
121.1 (7.5)
112.9 (5.6) 117.1 (4.5)
Height (cm)
32.9 (3.8)
32.2 (4.9)
32.0 (3.5)
32.1 (3.4)
32.5 (4.2)
32.7 (4.3)
31.4 (3.9)
27.6 (3.5)
25.5 (3.6)
22.2 (3.0)
19.8 (2.3)
19.2 (2.2)
16.7 (1.6)
14.3 (1.7) 15.5 (1.6)
LM (kg)
14.2 (5.5)
12.9 (3.4)
13.4 (3.8)
14.0 (4.5)
13.2 (5.4)
11.4 (4.7)
8.7 (3.9)
7.3 (3.4)
6.6 (3.2)
4.9 (2.1)
4.4 (1.8)
3.3 (1.6)
3.3 (1.7)
2.5 (1.1) 3.1 (1.6)
FM (kg)
2042.86 (354.43)
2010.35 (320.38)
1970.83 (299.99)
1964.39 (295.05)
1943.84 (312.03)
1862.99 (354.58)
1538.22 (282.12)
1333.02 (250.38)
1215.02 (219.09)
1013.65 (179.80)
909.36 (157.38)
862.83 (160.70)
745.02 (122.66)
599.45 (86.92) 664.06 (107.28)
BMC (g)
J Bone Miner Metab
123
J Bone Miner Metab
each gender and age group. For total body LM, significant gender differences were found in age groups 5–7 years (p \ 0.001), 8 years (p \ 0.01), 10 years (p \ 0.01) and 14–19 years (p \ 0.001). After 14 years, total body LM was significantly higher in boys than girls. However, for total body FM, the gender differences in these age groups were also observed in age groups 6 years (p \ 0.05), 8 years (p \ 0.01) 13 years (p \ 0.01), 14–18 years (p \ 0.001) and 19 years (p \ 0.01). After 13 years, total body FM was significantly higher in girls than boys. The relationship between body composition and bone mass was studied using the Pearson correlation coefficient and linear regression analysis. For all the subjects, significant positive correlations were observed between LM, FM and total body BMD, total body BMC, subcranial BMD, and subcranial BMC with r ranging from 0.451 to 0.977 (p \ 0.01). Bone mass for total body and subcranial skeleton as measured by BMC (g), BMD (g/cm2) and BMD/ height (g/cm3) was modeled as the dependent variable,
while age, height, LM and FM were treated as the independent variables. The model with the highest r2 value was chosen. Both LM and FM were consistent independent predictors of total body and subcranial bone mass in both sexes. The results of multiple linear regression identified LM as the stronger predictor of total body BMC and subcranial skeleton BMC while the FM contributed less as shown in Tables 2 and 3. For all the subjects, significant positive correlations were observed between total body LM, height, total body BMC and subcranial BMC (p \ 0.01). LM–height tended to be higher in boys than in girls. After evaluation of the different models, the exponential model was found to be the best fit for the association between LM and height and the linear model for total body and subcranial BMC and LM in both genders. The corresponding coefficients of determination (R2) were also calculated. Correlations between total body LM and height, and total body and subcranial BMC and LM are shown in Figs. 1 and 2,
Table 2 Multiple regression of total body and subcranial BMC, BMD, and BMD/height against age, height, LM and FM among boys Subcranial skeletona
Total body BMC
BMD
BMD/H
BMC
BMD
BMD/H
b
p
b
p
b
p
b
p
b
p
b
p
Age (years)
0.086
0.038
0.240
c
0.311
0.004
–
–
0.132
c
0.297
0.004
Height (cm)
0.494
c
0.239
c
-1.308
c
0.557
c
0.481
c
-0.537
c
LM (kg)
0.359
c
0.384
c
0.696
c
0.388
c
0.312
c
0.677
c
FM (kg)
0.064
c
0.062
0.002
0.088
0.012
0.060
c
0.071
c
0.172
c
model R2
0.892
0.753
0.279
0.901
0.873
0.343
BMC bone mineral content, BMD bone mineral density, BMD/H bone mineral density/height Total body with the head region of interest removed from analysis
a
c = p \ 0.001
Table 3 Multiple regression of total body and subcranial BMC, BMD, and BMD/height against age, height, LM and FM among girls Subcranial skeletona
Total body BMC
BMD
BMD/H
BMC
BMD
BMD/H
b
p
b
p
b
p
b
p
b
p
b
p
Age (years)
0.237
c
0.429
c
0.700
c
0.166
c
0.240
c
0.473
c
Height (cm) LM (kg)
–
–
-0.222
c
-1.901
c
0.061
0.032
–
–
-1.333
c
0.528
c
0.515
c
0.887
c
0.544
c
0.572
c
1.088
c
FM (kg)
0.259
c
0.220
c
0.370
c
0.257
c
0.178
c
0.355
c
model R2
0.932
0.796
0.415
0.843
BMC bone mineral content, BMD bone mineral density, BMD/H bone mineral density/height, a
Total body with the head region of interest removed from analysis
c = p \ 0.001
123
0.882
0.499
J Bone Miner Metab
Fig. 1 Correlations of total body lean mass calculated by DXA with height in boys (a) and girls (b)
Fig. 2 Correlations of total body and subcranial bone mineral content with lean mass calculated by DXA in boys (a) and girls (b)
respectively, for each gender group. The gender-specific LM-for-height percentile curves are displayed in Fig. 3. In general, the percentile curves for the two genders were
similar in shape. The LM-dependent percentile curves for BMC (Fig. 4) showed that BMC was closely associated with LM for total body and TBLH.
123
J Bone Miner Metab
Fig. 3 Lean mass percentile curves adjusted for height. Solid lines from the top represent the 97th, 75th, 50th, 25th, and 3rd percentiles
Discussion In this paper, we have produced age- and gender-specific normative body composition reference data for Chinese children and adolescents measured using a GE Lunar Prodigy DXA scanner. Furthermore, we have provided reference curves for LM–height and BMC–LM which will allow assessment of the skeletal and muscular status of Chinese children and adolescents using Crabtree’s approaches [17]. Ethnicity differences in body composition of children and adolescents has already been reported [21–23], proving that it is necessary to establish ethnicity-specific reference databases for the interpretation of pediatric DXA results. For example, Ellis et al. [22, 23] reported that LM was higher in black than in white boys and girls; no difference in LM was evident between white and Hispanic groups. Higher FM and body fat percentage values were observed in Hispanic boys and girls, even after correcting for body size. An increase in total body LM and FM with age is seen in our study, similar to that found in other studies [24–26]. Our data show that total body LM becomes significantly higher in boys than in girls from the age of 14 years, confirming previous results [26] and FM values become higher in girls than in boys from the age of 13 years, also in agreement with other studies [27]. In our literature review, puberty had a significant effect on body composition in boys and girls. For boys, greater LM may be caused by increased growth hormone and androgen concentrations. Synergism between growth hormone and androgens has already been described [28]. However, greater FM was discovered in girls after 13. It is generally accepted that body fat is a significant factor for female growth. Adipose
123
tissue may directly affect the menstrual cycle and ovulation, and hence fertility. Frisch and Revelle [29] suggested that a body fat of 17 % is needed to trigger the change in metabolic rate that leads to menarche, and a body fat of 22 % is required to maintain female reproductive ability [30]. In this study, participants were between the 3rd and 97th percentilea for height and weight on current standardized growth charts, roughly consistent with the mean values for Chinese children and adolescents. Therefore, we believe the data presented can represent LM and FM reference values and reflect the growth and development trends of a healthy population. In the present investigation, we adopted the method of Khosla et al. [31] to analyze the relationship between body composition and bone mass that is critically dependent on the specific parameter (i.e., BMC, BMD, BMD/height) in the analysis. Thus, using these correction factors tends to reduce the apparent influence of LM on BM. A number of body composition studies [14, 15, 17, 32–36] in children and adolescents have attempted to disentangle the independent relationships between LM and BM; their results are consistent with our own. Hogler W et al. [14] found that LM explained 96.8 % of the variation in total body BMC in boys and 95.1 % in girls. Pietrobelli et al. [32] showed that a 1 kg increase in LM was associated with a 0.04 kg increase in total body BMC on average, whereas a 1 kg increase in FM was associated with only a 0.009 kg increase in total body BMC. Arabi et al. [15], in a crosssectional study conducted with 363 school children aged 10–17 years, demonstrated that LM and FM are predictors of bone mass in boys and girls. The contribution of LM to BMC variance was larger than that of FM. Our results support the conclusion that both LM and FM are consistent independent predictors of total body and subcranial bone
J Bone Miner Metab
Fig. 4 Bone mineral content (BMC) percentile curves adjusted for lean mass. Solid lines from the top represent the 97th, 75th, 50th, 25th, and 3rd percentiles
mass in both sexes, even after adjustment for the wellknown predictors of BMC in healthy Chinese children and adolescents. Total body and subcranial BMC are more strongly associated with LM than with FM. Our data is in line with the notion that muscle development is correlated with skeletal development. All this suggests that analysis of bone parameters should be performed when considering muscle function. This study reports on the relationship between LM and height in subjects aged 5–19 years. Because muscle force is largely determined by body height [37, 38], muscle parameters should be measured in relationto body height during growth. Furthermore, it was found that LM values tend to plateau with age [22, 23, 26] but increase with increasing body height [12, 14, 17]. Our study reveals that LM values are positively and strongly related to height and
LM–height tends to be higher in boys than in girls, consistent with data published elsewhere [12, 14, 17]. In general, the LM assessment tools using DXA should be adjusted for height-. The International Society of Clinical Densitometry (ISCD) recommends that total body and TBLH BMD results should be adjusted for body size, using, for example, height, age or height-specific Z scores [2]. For this reason, height-dependent reference data for LM of the total body should be established for each gender group. Our study results indicate that total body and subcranial BMC and LM are highly correlated for each gender group, respectively, which is in line with the results from previous studies [12–14, 17, 39]. These statistical associations between LM and BMC both for total body and TBLH support the Frost mechanostat thesis that since muscle
123
J Bone Miner Metab
action delivers the largest loads and bone strains, reflecting the functional adaptation of bone to its function, then the increase in muscle development must precede and should determine the increase in BMC [39, 40]. Schoenau and Frost [41, 42] suggested that bone and muscle form a useful functional biological ‘‘muscle–bone unit’’ and that muscle– bone interactions during growth are key determinants of skeletal adaptation towards changing loads, which underlines the importance of evaluating bone and muscle tissues in the pathophysiology of childhood skeletal disorders. For example, LM assessment has been employed in studies of the muscular and skeletal status of children with idiopathic juvenile osteoporosis and osteogenesis imperfecta [43, 44], as well as by Hogler et al. in children with anorexia nervosa and growth hormone deficiency [14] and by Khadilkar in a female child presenting with stunting [13]. To evaluate skeletal status in children, the most accurate and reproducible skeletal sites for performing BMC and areal BMD measurements in this population are posteroanterior (PA) spine and TBLH as recommended by the ISCD [45]. Our study shows that subcranial BMC has a better correlation with LM than total body BMC and that subcranial skeleton parameters may be important indicators of muscular and skeletal status in children and adolescents. Pediatric body composition data are often difficult to interpret in children and adolescents. In a previous publication, a number of novel approaches have been proposed. Regarding application of the muscle–bone relationship, Scho¨nau et al. [39] suggested that measures of bone status (BMC) should be combined with those on muscle status (muscle cross-sectional area), using peripheral quantitative computed tomography (pQCT). The original idea was then extended by Ho¨gler et al. [14] to DXA which can differentiate of the origin of a low BMD or BMC/age, for example, short stature and primary, secondary, and mixed bone defects. With this in mind, Crabtree et al. has suggested another method for normalizing bone data [17], which shows how the associations between LM and height in addition to LM and BMC can be used to identify whether the primary abnormality is in muscle or bone. Data is analyzed in two steps: (1) the amount of LM relative to height is assessed (this reflects sacropenia); and (2) the amount of BMC relative to LM is assessed (reflecting the degree of osteopenia). In this study, we have adopted Crabtree et al.’s method [17] to develop percentile curves in addition to normative body composition data. The main perspective of our study is to provide new LM assessment tools helpful for elucidating the origin of low bone mass in children and adolescents, focusing on the functional biological ‘‘muscle–bone unit’’. We acknowledge some potential limitations to this study. The primary limitation is that it was a cross-sectional study and further longitudinal data need to be
123
obtained. Secondly, there are other limitations concerning DXA. Due to differences in scan modes, software version [46], and the calibration methods adopted by different DXA manufacturers [47, 48], variations between different DXA scanners are known to result in substantial differences in body composition measurements. Our reference data is limited only to results derived from the Lunar Prodigy DXA densitometer; these reference curves cannot be used with the results from other brands of DXA scanner. The final limitation is that we could not acquire the Tanner stage of the subjects. Furthermore, caution should be advised when attempting to utilize the provided normative values due to the differences in genetic background, nutritional habits, physical activity level, and other lifestyle choices of the study population. In summary, we have presented an ethnicity- and gender-specific body composition normal reference database as well as percentile curves for normalization of muscle– bone data by DXA for Chinese children and adolescents aged 5–19 years. To the best of our knowledge, these are the first total body normal reference data for Chinese children and adolescents. Therefore, it is worthwhile establishing the new LM assessment reference values allowing precise comparison with the given results and then calculating the Z score in Chinese children and adolescents. The results of this study may be used for clinical differentiation and assessment of children with different aetiologies of pediatric bone disorders. Further work will be required to evaluate this relationship with other clinical conditions in China as measured by the Lunar Prodigy DXA densitometer. Acknowledgments The authors would like to express their gratitude to all participating children and their parents. We are grateful to Dr. Qi Zhou, GE Healthcare Shanghai and Dr. Jing Xiang, First Hospital of Jiaxing for their useful comments and suggestions. We also thank the staff members of the Department of Nuclear Medicine, First Affiliated Hospital of Jinan University for excellent technical support. Conflict of interest None of the authors have any personal or financial conflicts of interest.
References 1. Bachrach LK (2005) Osteoporosis and measurement of bone mass in children and adolescents. Endocrinol Metab Clin North Am 34:521–535 2. Gordon CM, Bachrach LK, Carpenter TO, Crabtree N, El-Hajj FG, Kutilek S, Lorenc RS, Tosi LL, Ward KA, Ward LM, Kalkwarf HJ (2008) Dual energy X-ray absorptiometry interpretation and reporting in children and adolescents: The 2007 ISCD Pediatric Official Positions. J Clin Densitom 11:43–58 3. Fewtrell MS (2003) Bone densitometry in children assessed by dual X-ray absorptiometry: uses and pitfalls. Arch Dis Child 88:795–798
J Bone Miner Metab 4. Molgaard C, Thomsen BL, Prentice A, Cole TJ, Michaelsen KF (1997) Whole body bone mineral content in healthy children and adolescents. Arch Dis Child 76:9–15 5. Guo B, Xu Y, Gong J, Tang Y, Xu H (2013) Age trends of bone mineral density and percentile curves in healthy Chinese children and adolescents. J Bone Miner Metab 31:304–314 6. Binkovitz LA, Henwood MJ, Sparke P (2008) Pediatric DXA: technique, interpretation and clinical applications. Pediatr Radiol 38:S227–S239 7. Binkley TL, Berry R, Specker BL (2008) Methods for measurement of pediatric bone. Rev Endocr Metab Disord 9:95–106 8. Kim J, Wang Z, Heymsfield SB, Baumgartner RN, Gallagher D (2002) Total-body skeletal muscle mass: estimation by a new dual-energy X-ray absorptiometry method. Am J Clin Nutr 76:378–383 9. Arabi A, Nabulsi M, Maalouf J, Choucair M, Khalife H, Vieth R, El-Hajj Fuleihan G (2004) Bone mineral density by age, gender, pubertal stages, and socioeconomic status in healthy Lebanese children and adolescents. Bone 35:1169–1179 10. Hogler W, Briody J, Woodhead HJ, Chan A, Cowell CT (2003) Importance of lean mass in the interpretation of total body densitometry in children and adolescents. J Pediatr 143:81–88 11. Arabi A, Tamim H, Nabulsi M, Maalouf J, Khalife H, Choucair M, Vieth R, El-Hajj Fuleihan G (2004) Sex differences in the effect of body-composition variables on bone mass in healthy children and adolescents. Am J Clin Nutr 80:1428–1435 12. Ogle GD, Allen JR, Humphries IR, Lu PW, Briody JN, Morley K, Howman-Giles R, Cowell CT (1995) Body-composition assessment by dual-energy X-ray absorptiometry in subjects aged 4–26 years. Am J Clin Nutr 61:746–753 13. Ferretti JL, Capozza RF, Cointry GR, Garcia SL, Plotkin H, Alvarez FML, Zanchetta JR (1998) Gender-related differences in the relationship between densitometric values of whole-body bone mineral content and lean body mass in humans between 2 and 87 years of age. Bone 22:683–690 14. Pludowski P, Matusik H, Olszaniecka M, Lebiedowski M, Lorenc RS (2005) Reference values for the indicators of skeletal and muscular status of healthy Polish children. J Clin Densitom 8:164–177 15. Khadilkar AV, Sanwalka NJ, Chiplonkar SA, Khadilkar VV, Mughal MZ (2011) Normative data and percentile curves for dual energy X-ray absorptiometry in healthy Indian girls and boys aged 5-17 years. Bone 48:810–819 16. Frost HM, Schonau E (2000) The ‘‘muscle-bone unit’’ in children and adolescents: a 2000 overview. J Pediatr Endocrinol Metab 13:571–590 17. Crabtree NJ, Kibirige MS, Fordham JN, Banks LM, Muntoni F, Chinn D, Boivin CM, Shaw NJ (2004) The relationship between lean body mass and bone mineral content in paediatric health and disease. Bone 35:965–972 18. Li H, Ji CY, Zong XN, Zhang YQ (2009) Height and weight standardized growth charts for Chinese children and adolescents aged 0 to 18 years. Zhonghua Er Ke Za Zhi 47:487–492 19. Ministry of Education of the People’s Republic of China (2007) Report on the physical fitness and health surveillance of Chinese school students. Higher Education Press, Beijing 20. Cole TJ, Green PJ (1992) Smoothing reference centile curves: the LMS method and penalized likelihood. Stat Med 11:1305–1319 21. Nelson DA, Barondess DA (1997) Whole body bone, fat and lean mass in children: comparison of three ethnic groups. Am J Phys Anthropol 103:157–162 22. Ellis KJ (1997) Body composition of a young, multiethnic, male population. Am J Clin Nutr 66:1323–1331 23. Ellis KJ, Abrams SA, Wong WW (1997) Body composition of a young, multiethnic female population. Am J Clin Nutr 65:724–731
24. Sala A, Webber CE, Morrison J, Beaumont LF, Barr RD (2007) Whole-body bone mineral content, lean body mass, and fat mass measured by dual-energy X-ray absorptiometry in a population of normal Canadian children and adolescents. Can Assoc Radiol J 58:46–52 25. Alwis G, Rosengren B, Stenevi-Lundgren S, Duppe H, Sernbo I, Karlsson MK (2010) Normative dual energy X-ray absorptiometry data in Swedish children and adolescents. Acta Paediatr 99:1091–1099 26. Boot AM, Bouquet J, de Ridder MA, Krenning EP, de Muinck Keizer-Schrama SM (1997) Determinants of body composition measured by dual-energy X-ray absorptiometry in Dutch children and adolescents. Am J Clin Nutr 66:232–238 27. Kim K, Yun SH, Jang MJ, Oh KW (2013) Body fat percentile curves for Korean children and adolescents: a data from the Korea national health and nutrition examination survey 2009-2010. J Korean Med Sci 28:443–449 28. Martin LG, Grossman MS, Connor TB, Levitsky LL, Clark JW, Camitta FD (1979) Effect of androgen on growth hormone secretion and growth in boys with short stature. Acta Endocrinol 91:201–212 29. Frisch RE, Revelle R (1970) Height and weight at menarche and a hypothesis of critical body weights and adolescent events. Science 169:397–399 30. Frisch RE (1987) Body fat, menarche, fitness and fertility. Hum Reprod 2:521–533 31. Khosla S, Atkinson EJ, Riggs BL, Melton LJ 3rd (1996) Relationship between body composition and bone mass in women. J Bone Miner Res 11:857–863 32. Pietrobelli A, Faith MS, Wang J, Brambilla P, Chiumello G, Heymsfield SB (2002) Association of lean tissue and fat mass with bone mineral content in children and adolescents. Obes Res 10:56–60 33. Young D, Hopper JL, Nowson CA, Green RM, Sherwin AJ, Kaymakci B, Smid M, Guest CS, Larkins RG, Wark JD (1995) Determinants of bone mass in 10- to 26-year-old females: a twin study. J Bone Miner Res 10:558–567 34. Young D, Hopper JL, Macinnis RJ, Nowson CA, Hoang NH, Wark JD (2001) Changes in body composition as determinants of longitudinal changes in bone mineral measures in 8 to 26-yearold female twins. Osteoporos Int 12:506–515 35. Ilich JZ, Skugor M, Hangartner T, Baoshe A, Matkovic V (1998) Relation of nutrition, body composition and physical activity to skeletal development: a cross-sectional study in preadolescent females. J Am Coll Nutr 17:136–147 36. Valdimarsson O, Kristinsson JO, Stefansson SO, Valdimarsson S, Sigurdsson G (1999) Lean mass and physical activity as predictors of bone mineral density in 16-20-year old women. J Intern Med 245:489–496 37. Round JM, Jones DA, Honour JW, Nevill AM (1999) Hormonal factors in the development of differences in strength between boys and girls during adolescence: a longitudinal study. Ann Hum Biol 26:49–62 38. Parker DF, Round JM, Sacco P, Jones DA (1990) A cross-sectional survey of upper and lower limb strength in boys and girls during childhood and adolescence. Ann Hum Biol 17:199–211 39. Schoenau E, Neu CM, Beck B, Manz F, Rauch F (2002) Bone mineral content per muscle cross-sectional area as an index of the functional muscle-bone unit. J Bone Miner Res 17:1095–1101 40. Schoenau E (2005) From mechanostat theory to development of the ‘‘Functional Muscle-Bone-Unit’’. J Musculoskelet Neuronal Interact 5:232–238 41. Schoenau E, Frost HM (2002) The ‘‘muscle-bone unit’’ in children and adolescents. Calcif Tissue Int 70:405–407 42. Fricke O, Schoenau E (2007) The ‘Functional Muscle-Bone Unit’: probing the relevance of mechanical signals for bone
123
J Bone Miner Metab development in children and adolescents. Growth Horm IGF Res 17:1–9 43. Pludowski P, Lebiedowski M, Olszaniecka M, Marowska J, Matusik H, Lorenc RS (2006) Idiopathic juvenile osteoporosis–– an analysis of the muscle-bone relationship. Osteoporos Int 17:1681–1690 44. Pludowski P, Matusik HLRS (2003) Evaluation of musculoskeletal system in OI and IJO subjects. Osteoporosis Int 14:S31– S32 45. Khan AA, Bachrach L, Brown JP, Hanley DA, Josse RG, Kendler DL, Leib ES, Lentle BC, Leslie WD, Lewiecki EM, Miller PD, Nicholson RL, O’Brien C, Olszynski WP, Theriault MY, Watts NB (2004) Canadian panel of the international society of clinical
123
densitometry standards and guidelines for performing central dual-energy X-ray absorptiometry in premenopausal women, men, and children. J Clin Densitom 7:51–64 46. Laskey MA, Prentice A (1999) Comparison of adult and paediatric spine and whole body software for the lunar dual energy X-ray absorptiometer. Br J Radiol 72:967–976 47. Tothill P, Hannan WJ (2002) Bone mineral and soft tissue measurements by dual-energy X-ray absorptiometry during growth. Bone 31:492–496 48. Tothill P, Avenell A, Love J, Reid DM (1994) Comparisons between hologic, lunar and norland dual-energy X-ray absorptiometers and other techniques used for whole-body soft tissue measurements. Eur J Clin Nutr 48:781–794
