Puberty: Physiology and Abnormalities, 1st ed. 2016

6. Bone Development in Children and Adolescents

Mihail A. Boyanov 


Department of Internal Medicine, Clinic of Endocrinology and Metabolism, University Hospital Alexandrovska, Medical University Sofia, 1, Georgi Sofiyski Street, 1431 Sofia, Bulgaria

Mihail A. Boyanov



Bone developmentGrowth plateBone cellsBone ageSex steroidsGrowth factorsBone mineral density


Bone is a living tissue and integrated into the compact skeleton it has to fulfill a number of vital functions: (1) to provide a mechanical barrier for the protection of soft tissues and internal organs; (2) to allow movement, together with muscles (the so-called muscle–bone unit); (3) to store calcium and other molecules involved in the equilibrium of the extracellular space; and (4) to protect the bone marrow and to interact with its cell lines, as well as many other functions. Bone development during childhood and adolescence is a process of great complexity and can be viewed from many aspects [13]. Among the most debated ones are the development and maintenance of the growth plate, the bone mineral accrual at the different skeletal parts, and the factors contributing to the regulation of all these events. The implementation of bone age is another interesting point combining classical clinical use with up-to-date computerized applications. These aspects will be commented below.

Mechanisms of Bone Development: The Role and Structure of the Growth Plate

The bone tissue is derived from different parts of the mesoderm. Craniofacial elements develop through the process of intramembranous ossification, while long bones are the product of endochondral ossification. In the latter type of ossification, the mesodermal stem cells form condensed groups and differentiate into chondrocytes producing an extracellular matrix, rich in collagen type 2. The chondrocytes in the midportion become hypertrophic and start to secrete collagen type 10, which can be then mineralized. Cells from the perichondrium can differentiate into osteoblasts providing the basis for the so-called bone collar—the initiation site for the cortical envelope. A critical step in this process is the invasion by blood vessels, followed by apoptosis of terminally differentiated chondrocytes and resorption of the calcified cartilage matrix. The cartilage is replaced by bone and vascular elements, building the primary ossification center.

The remaining chondrocytes at both ends of the long bones provide the basis for further bone growth—the growth plate. The growth plate is entrapped between the epiphyseal and metaphyseal bone and consists of distinct layers with different functions. The most superficial layer (adjacent to the epiphysis) is built by round chondrocytes (resting zone), followed by more mature and strongly proliferating columnar ones (proliferation zone). They further mature to the so-called hypertrophic chondrocytes (which build the differentiation zone). The chondrocytes adjacent to the metaphysis are subjected to apoptosis and cell death. The junction between the growth plate and the metaphysis is the site for invasion of new vessels, osteoclasts, and osteoblasts. The osteoclasts resorb the calcified cartilage matrix, while the osteoblasts deposit mineralized bone matrix on the remnants. So, the chondrocytes in the growth plate are both spatially and temporarily differentiated. The linear bone growth, in its turn, leads to an age gradient in the metaphysis [4]. The newest part of the metaphysis is adjacent to the growth plate, while the oldest one—to the epiphysis. During childhood, the growth plate matures and it fuses (closes) at the end of puberty under the influence of gonadal steroids and other specific factors. A very comprehensive depiction of these processes might be found elsewhere [13].

Modern molecular biology allows a closer look into the mechanisms of bone development and the key regulating factors. A number of questions concerning the growth plate maturation and fusion, as well as growth factors and local regulators, are still being debated. Most data come from mice. A very comprehensive picture of all these factors is provided in the review by Emons et al. [5]. Chondrocytes receive a variety of signals from the perichondrial cells, including bone morphogenetic proteins (BMPs) , fibroblast growth factors (FGFs), and other highly conserved secreted signaling molecules that regulate cell-to-cell interactions during development and adult tissue homeostasis such as Wnt signaling. BMP signaling is essential for converting the mesenchymal cells into chondrocytes, as well as in the later stages of cartilage development. BMPs have distinct and sometimes opposing effects in the different chondrocytic zones (agonistic in the hypertrophic and antagonistic in the resting and proliferative zones) [6]. The Wnt signaling pathway is a conserved pathway in animals: the name Wnt being resultant from a fusion of the name of the Drosophila segment polarity gene wingless and the name of the vertebrate homolog integrated or int-1.

FGF-18 and FGF-9 are considered important in regulating chondrogenesis [78]. FGFs may act as antagonists to BMPs. Activating mutations in FGF-3 receptor may lead to achondroplasia or hypochondroplasia, while loss-of-function mutations in the responsible gene are accompanied by tall stature in humans [910]. The activation of the canonical Wnt pathway results in activation of β-catenin and induction of osteoblast, instead of chondrocyte formation [11]. A number of transcription factors (such as Runx2, Sox9, and others) and growth factors (such as transforming growth factor-beta, TGF-β) are also implicated in the complex regulatory processes. Vascular endothelial growth factor (VEGF) is a potent stimulator of blood vessel invasion and expansion. It is also considered crucial for the estrogen-driven growth plate fusion. A study in rats found that the expression of VEGF in the growth plate was stimulated by estradiol, although the precise role of VEGF in estrogen-mediated growth plate fusion could not be demonstrated [12]. Key factors in the regulation of the growth plate are also the hormonal and paracrine factors, such as parathyroid hormone (PTH), parathyroid hormone-related peptide (PTHrp) , vitamin D, and Indian hedgehog (Ihh). Ihh and PTHrp form a negative feedback loop regulating the hypertrophic differentiation of chondrocytes. Ihh, produced by the early hypertrophic chondrocytes, stimulates the expression of PTHrp, which in turn slows down the differentiation and keeps the chondrocytes in the proliferative stage [13]. A homozygous mutation in the Ihh gene may lead to disproportional short stature, brachydactyly, and premature fusion of the growth plates [14]. An inactivating mutation in the PTH receptor may lead to chondrodysplasia with advanced bone maturation (Blomstrand chondrodysplasia), while an activating mutation leads to short stature and delay in bone maturation in Jansen chondrodysplasia [1516]. Vitamin D deficiency is known to result in increased width of the hypertrophic chondrocytic zone due to decreased cell death and in delayed invasion by angiogenic and bone cells [17]. All these hormonal, transcriptional, and other regulating factors have their origins in distinct patterns of gene expression. A recent study, looking for evolutionary conserved networks, identified a number of growth pathways (Notch, VEGF, TGF-β, Wnt, and glucocorticoid receptor) [18].

The classical model of growth plate maturation and epiphyseal fusion involves the sex steroids—estrogens and androgens. It is derived from the observation that precocious puberty is accompanied by early skeletal maturation and early growth stop, while delayed puberty or hypogonadism leads to slow skeletal maturation and excessive growth of the long bones. There was active debate on the relative contributions of estrogens and androgens in both sexes. A role has been clearly attributed to the estrogen receptor-α (ER-α), while the role of the estrogen receptor-β (ER-β) remains to be elucidated [1920]. Estrogen is thought to accelerate growth plate senescence (maturation and aging), in addition to a genetically programmed mechanism intrinsic to the growth plate itself [21]. Estrogens can also be produced locally in the growth plate. Androgens were thought to act mainly via conversion into estrogens, but recent work has shown an independent pathway for androgenic substances, since androgen receptors are found in the human growth plate, as well as on human osteoblasts [2223].

While sexual maturation and puberty progress, the growth plate is a subject of structural and functional changes—the so-called senescence. It is partly due to genetic programming and finite proliferative capacity. The classical hypothesis is that of apoptosis resulting in cell shrinkage, DNA fragmentation, and degradation of the cytoplasm and nucleus in the absence of signs of inflammation. Other suggested mechanisms include autophagy, trans-differentiation of hypertrophic chondrocytes into osteoblasts (the oldest hypothesis), hypoxia, and others. However, human studies focused on the growth plate have not provided conclusive evidence for all these hypotheses [24].

Osteoclastogenesis and Osteoblastogenesis

Two cell types are indispensable for the development of mature bone tissue—osteoblasts and osteoclasts. Once the hypertrophic cartilage is invaded by blood vessels and VEGF is secreted, it is the osteoclasts that resorb cellular and matrix debris to free space for the newly built bone. Osteoblasts, in their turn, react to signals such as BMPs, TGF-β, and others and start building collagen and non-collagen proteins, which further become mineralized. A number of regulating factors are key players for osteoclast recruitment, differentiation, and survival. Both macrophages and osteoclast precursor cells originate from a common myeloid precursor under the action of transcription factor PU [25]. This factor also enhances the transcription of the receptor for another crucial regulator, the so-called receptor activator of nuclear factor-k-β (RANK) [26]. The ligand of this receptor (RANKL), together with the macrophage colony-stimulating factor (M-CSF) , is critical for osteoclastogenesis. The M-CSF acts in the earlier stages of osteoclastogenesis, while the RANKL is responsible for osteoclast recruitment, differentiation, maturation, and survival [27]. The activation of the nuclear factor-k-β by the binding of RANKL to the membrane-bound receptor leads to c-fos expression and induction of osteoclast-specific genes [28]. RANKL is secreted by the osteoblasts in response to different cellular signals. RANKL is regulated negatively by a soluble decoy receptor called osteoprotegerin (OPG), which is also secreted by the osteoblasts [2930]. Thus, one of the key regulating cells for the osteoclastogenesis and survival is the osteoblast itself. The osteoblast is the primary target for different signals to the bone in physiological and pathological conditions, such as vitamin D, parathyroid hormone, PTHrp, different cytokines—interleukins (in inflammatory and immune diseases)—and many others [3133]. It is interesting to note that the osteoclast gives signals back to the osteoblast, thus modulating the equilibrium between RANKL and OPG. Among these are different BMPs, TGF-β, and others [34]. The discovery of RANKL led to the development of a targeted, fully human antibody, used in the treatment of postmenopausal osteoporosis [35]. The scientific work in the field of postmenopausal osteoporosis led to the discovery of the third important cell type in the development of bone. Osteocytes, previously thought to be aging and exhausted osteoblasts, play an important role in the suppression of bone formation and the induction of resorption. Through their dendritic processes, they sense mechanical forces, communicate with each other, and send signals to the bone-forming cells, such as the inhibitor of the canonical Wnt pathway—osteosclerostin (SOST) [3637]. SOST binds two synergistically acting receptor families and stabilizes the so-called β-catenin protein, which regulates the expression of downstream target genes [38]. The homozygous loss-of-function mutation in SOST may lead to sclerosteosis, while mutations, affecting its transcription, cause van Buchem disease—both associated with high bone mass phenotypes [39]. The discovery of SOST gave way to the development of antagonistic antibodies, which promise to become one of the most potent bone-forming agents under study [40].

At the structural level, bone development during childhood and adolescence is the net result of two processes—bone modeling and bone remodeling [41]. The evolutionary role of bone modeling is to build new bone, while during bone remodeling, damaged bone is first removed and the resorption cavities are subsequently mineralized [1]. Bone modeling is typical for the growing bone and results in bone mineral apposition and bone size increases. During bone modeling, osteoclasts and osteoblasts are uncoupled; they act separately at different sites of the bone. Bone modeling is most active until skeletal maturity and was believed to stop at skeletal maturity, although recent work has shown some modeling to occur at later ages at a very modest rate [42]. Bone remodeling is characterized by the coupling of osteoblasts and osteoclasts, which build the so-called bone remodeling unit (BMU) [43]. The osteoclasts first excavate a resorption lacuna, which is then filled by the osteoblasts and subjected to a slow process of mineralization. The frequency of activation of the BMU and the depth of the resorption cavities are among the key factors determining the bone loss in adulthood and senescence [44].

During childhood and adolescence, bones grow in both length and width. Periosteal apposition is responsible for the growth in width, while endochondral ossification for the growth in length. Resorption occurs inside the bone, while new bone is deposited on the outer surface (subperiosteal apposition driven mainly by osteoblasts and bone modeling). Bones are subject to constant changes in shape and size. This process terminates around the time of skeletal maturity with the bone resorption continuing throughout life at different rates according to age [145].

Bone Mineral Accretion in Children and Adolescents

The question of bone mineral accrual during adolescence and of the timing of peak bone mass is very difficult to address. Firstly, bone mass can be measured at different sites—the vertebrae, femur, radius, tibia, and whole body—which grow and develop at different rates and time scales. The different skeletal parts are affected by different factors (e.g., weight-bearing activities). Secondly, there is a great variety of measuring techniques—dual-energy X-ray absorptiometry (DXA), quantitative computed tomography (QCT), high-resolution computed tomography, and magnetic resonance imaging (hr-pQCT, hr-MRI). The most widely used technique, DXA, provides data on bone mineral content (BMC) and bone mineral density (BMD) . DXA-derived BMD is bidimensional; it is called areal BMD and does not properly reflect the contribution of the third dimension (bone depth or the anteroposterior diameter). BMC and aBMD change rapidly during growth because of bone size increases. An attempt to overcome this intrinsic methodological problem is to measure the true volumetric BMD by QCT or its peripheral modification, pQCT. Volumetric BMD can also be calculated if bone density scans in two planes are available, such as vBMD of the vertebrae derived from anteroposterior and lateral spine scans. In a previous study, an algorithm for calculation of forearm vBMD has been proposed, although the accuracy of all calculated parameters is debatable [46]. The third problem is due to the different proportions of cortical and trabecular bone in the skeletal parts. Cortical and trabecular bone might be regarded as two distinct components and their interaction determines the net effect on bone development. In an experimental model, calculated cortical and trabecular forearm BMD showed a distinct pattern of changes over time in adult women [47]. Only the QCT, pQCT, and high-resolution techniques are capable of distinguishing both components of the bone. And fourth, a large number of factors are affecting bone growth and development, such as heredity, age, sex, race, physical activity, nutrition, pubertal stage, etc. It is almost impossible to dissect a single factor and define its contribution to the whole. To summarize, the problem of different skeletal sites, measurement techniques, and biological factors should be kept in mind, when interpreting data about bone mineral changes during childhood and adolescence.

A DXA-based study in healthy Italian children showed that before puberty, boys and girls had equal BMD at the lumbar spine, while the femoral neck BMD was higher in boys [48]. In the areas rich of cortical bone, the boys showed higher values before puberty, while trabecular bone was equivalent until the age of 9 and increasing more steeply in girls thereafter. The authors concluded that male sex and lean mass were predictors for the higher cortical bone mass in boys, while female sex and pubertal stage predicted the higher trabecular bone accrual in girls [48]. Another DXA-based study explained the differences in BMC as mainly due to the differences in body size [49]. Body size was held responsible for most of the racial/ethnic differences in BMC. In a cohort of Spanish adolescents, females had higher values of the DXA-derived BMC and BMD at most of the sites, probably due to the smaller diameters of bones [50]. Bone growth from 11 to 17 years was studied by pQCT at the distal forearm [51]. This study showed that bone width and mineral content increased with age albeit independently, which resulted in a modest decrease of vBMD during early puberty followed by a rapid increase. Physical growth occurred at a higher tempo than bone growth. By the age of 17, boys had attained 86 % of the reference adult BMC and vBMD, while girls had attained 93 and 94 %, respectively [51]. A very interesting study assessed the structural and biomechanical basis for sexual dimorphism at the hip by using the DXA-derived FN areal BMD and measured the periosteal diameter to estimate the endocortical diameter, cortical thickness, section modulus (a measure of bending strength), and buckling ratio (indices for structural stability) [45]. In this study, FN cortical thickness or volumetric density did not differ in young adult women and men after adjustment for height and weight. The sex differences in geometry were best described by the further displacement of the cortex from the FN neutral axis in young men, which produced a 13.4 % greater bending strength than in young women. The changes in FN diameters, cortical thickness, and geometry during puberty are presented in Fig. 6.1 according to [45].


Fig. 6.1

Changes in FN diameters, cortical thickness, and geometry during puberty. Cortical thickness or volumetric density does not differ in young adult women and men; the sex differences in geometry are best described by the further displacement of the cortex from the FN neutral axis in young men [45]. This is now thought to be partly due to the longer prepubertal growth and intrapubertal growth in males than females (Courtesy of Prof. E. Seeman)

The effects of sex, race, and puberty on cortical bone were examined by pQCT of the distal tibia [52]. This study found higher cortical measures in blacks than whites in Tanner stages 1–4; however, differences were negligible in Tanner stage 5. Cortical BMC, periosteal and endosteal circumferences, and section modulus were higher in pubertal males than females; however, cortical BMD was higher in Tanner 3–5 females. The authors were unable to explain maturation-specific differences in cortical BMD and dimensions solely by differences in bone length or muscle [52]. Cortical porosity at the distal radius and tibia was studied during pubertal growth by hr-pQCT [53]. At the radius, girls had higher cortical density and lower cortical porosity than boys, while boys had higher trabecular bone volume ratios and larger cortical cross-sectional areas. These results could provide a solid basis for the sex- and maturity-related differences in bone microarchitecture and strength [53].

The task of describing bone changes is further complicated by the contribution of muscle development to the so-called muscle–bone unit. Bone and muscle properties were studied prospectively by tibial DXA and pQCT—the growth velocity of the muscle cross-sectional area peaked earlier than tibial BMC and later than tibial outer dimensions [54]. Tibial length was the first to stop increasing 2 years after menarche, while all other muscle and bone parameters continued to increase even at the age of 18 years. This study corroborated the hypothesis that the development of lean mass precedes that of bone mass but did not support the role of the muscles as the central driving force for bone growth during puberty [54]. In summary, all the data coming from bone measurement techniques show an absence of gender-specific differences in bone mass until the onset of puberty. The bone maturation period seems longer in males than females, and the increase in bone size is greater in males. Changes in vBMD between sexes are negligible at the end of puberty; the differences in BMD are due mainly to the differences in bone size and architecture, as stated by other authors [155].

Peak bone mass (PBM) is essential for the future risk of osteoporosis and fragility fractures. The exact age, at which PBM is attained, remains a matter of debate. A DXA-based study found that in males, bone mass at different skeletal sites continued to increase between 15 and 18 years, while in girls, it slowed down at the levels of both lumbar spine and femoral neck (FN) at 15–16 years of age [56]. Girls attained values near PBM at an earlier age than boys. In the 14–15-year-old female group, BMD in L2–L4, FN, and femoral shaft corresponded to 99.2, 105.1, and 94.1 %, respectively, and BMC in L2–L4 to 97.6 % of the mean values for 20–35-year-old women [56]. In a similar study, the contribution of the third decade of life to BMD in women was found to be only 6.8 % at the LS and 4.8 %—at the forearm [57]. The authors located the end of bone mass acquisition around the age of 28.3 and 29.5 years. The sex differences in bone mass acquisition during growth were estimated in the Fels Longitudinal Study [58]. This study showed that PBM and density are attained generally between the ages of 20 and 25 years and occur earlier in females than males. vBMD was shown to rise much slower than BMC and aBMD, due to the different contributions of bone size and density [59]. Peak vBMD of the lumbar vertebrae was reached between 22 and 29 years of age, while peak values were reached substantially earlier for the femoral neck (around 12 years of age) and ultradistal radius (around 19 years of age) [59].

Factors Contributing to Bone Development in Children and Adolescents

Bone development occurs under the influence of many contributing factors—age, sex, race, pubertal development, body weight and height (size), physical activity, nutrition, and many others. Age, sex, and pubertal stage also affect the skeleton by supporting a specific hormonal milieu. Some of these factors have been extensively studied in the last decades.

The influence of being overweight was studied in 11–13-year-old boys by DXA at the LS, FN, and whole body (WB) [60]. Overweight boys displayed similar values for LS and WB and lower for the FN, compared to normal-weight controls . In normal-weight boys, fat-free mass was the major determinant of bone mineral indices, while in overweight boys, it correlated only with the FN; the LS and WB BMD correlating better with the fat mass [60]. This study showed that the influence of body weight on BMD is not uniform and cannot be explained as putting more mechanical stress on the skeleton and thus stimulating the growth of the muscle–bone unit. Another study tested the association of lean and fat body mass with bone mass during pre- and midpuberty and found appendicular lean mass to be the strongest determinant [61]. These associations were also studied by tibial pQCT in the Avon Longitudinal Study of Parents and Children [62]. Lean mass showed a positive correlation with cortical BMC in both boys and girls and its correlation with periosteal circumference was stronger in girls. Fat mass showed a stronger correlation with cortical BMC and periosteal circumference and a negative one with endosteal circumference in girls. The authors concluded that lean mass stimulated the development of cortical bone mass in a similar way in boys and girls, while fat mass was a stronger stimulus only in girls [62]. The difficulty in differentiating the relative contributions of lean and fat mass to the BMD values was confirmed in adult women also (see Table 6.1) [63].

Table 6.1

The correlation coefficients (R-square) from curve estimation analyses with lumbar spine BMD (as the dependent variable) and body weight, % body fat, and fat and lean mass (as the independent variables) are in the same magnitude




Fat mass

Lean mass

Total body weight

% Body fat














































The significance was p < 0.001 for all equations including body weight, fat, and lean mass and 0.01 > p > 0.001 for the % of body fat (Reprinted from Boyanov M. Body fat, lean mass and bone density of the spine and forearm in women. Central European Journal of Medicine 2014;9(1):121–125. With permission from Springer Science + Business Media)

Physical activity is a crucial activator of skeletal modeling and remodeling. The effect of physical activity early in life was tested at ages 5, 13, and 15 years in the Iowa Bone Development Study [64]. In boys, physical activity predicted later spine BMC, while this was not the case in girls. In another longitudinal study in boys, vigorous physical activity and sedentary time had a significant effect only on FN BMD [65]. The negative influence of sedentary time on bone mineral parameters was tested prospectively in peripubertal boys and was confirmed for all areas of interest (WB, LS, FN) with the strongest negative effect on FN BMD [66]. The type of sport or vigorous physical activity should also be kept in mind. The influence of different sports on bone mass was studied in girls [67]. In the pubertal group, arms BMD, pelvis BMD, and FN BMD were higher in soccer and handball players than in the control group, while swimmers had significantly higher values in the arms BMD. The authors recommended sport activities that support body weight as an important factor for achieving an optimal peak bone mass [67]. A meta-analysis of the effect of exercise on pediatric bone and fat revealed a small positive effect of bone-targeted exercise on WB, FN, LS BMC, and fat mass [68]. Another meta-analysis examined the effects of weight-bearing exercise in girls and found that these were site specific and affected primarily LS BMD, while BMC was increased at both LS and FN [69]. The authors also noted that physical activity for more than 3 days per week resulted in significantly greater values [69]. All these studies support the beneficial effect of exercise on bone development and health. The importance of mechanical influences for bone development in children has led to the formulation of a “mechanostat paradigm” [70].

Nutrition is another important aspect for optimal bone health. International and national agencies have adopted recommendations for optimal daily calcium and vitamin D intake at different ages. An epidemiological study in women aged 20–49 years revealed that milk consumption during childhood and adolescence was correlated with bone density and fracture risk later in life [71]. The positive association of calcium intake with bone mass was displayed in another study in adolescents [72]. The effect of calcium supplementation was more pronounced in cortical-rich (radius and femur) than in trabecular-rich bone (LS) [72]. An interventional study in 8-year-old prepubertal girls showed an increase in bone mass by 0.25 standard deviations only in those with very low-calcium intake at baseline [73]. A more recent study examined the interaction of low-calcium diet and low 25(OH)D levels in late-pubertal girls and found that LS BMC and BMD were not associated with 25(OH)D, except when calcium intake was below 600 mg/day [74]. The response to calcium intake interacts also with physical activity: the higher the calcium intake, the more pronounced is the effect of physical activity on bone growth [7576]. In two other studies (a cross-sectional and longitudinal one) the effect of physical activity on WB BMC accrual varied only among Tanner stages after adjustment for calcium intake [7778]. The positive interaction of high-protein intake and physical activity on bone strength and microstructure from prepuberty to mid-to-late adolescence was shown in a study using hr-pQCT of the tibia [79]. Bone mass accrual might be influenced also by harmful habits, such as early initiation of smoking and alcohol drinking, as shown in a cohort study in late adolescent girls [80].

Puberty is one of the main driving forces of bone development. The timing of puberty was shown to be inversely associated with peak bone mass, e.g., individuals with late puberty had lower bone mass in young adulthood [8182]. In Tanner stage 5, height growth was found to exert a more pronounced effect on bone accrual than at puberty start [8182]. It is intuitive to make the assumption that puberty acts on bone through a variety of mechanisms, among which the hormonal ones play a leading role [8384]. However, it is very difficult to dissect (discern) the relative contribution of every single hormone or factor. BMD, bone markers, and gonadal steroids were studied in girls and boys in different pubertal stages [85]. BMD significantly increased until Tanner stage 4 in girls, while in boys at Tanner stage 4 , it was higher than in all other pubertal stages. There was a modest correlation of BMD with testosterone (T) and estrogen (E) in boys and only with estrogen in girls. Bone alkaline phosphatase was a better predictor of bone mass in girls than in boys [85]. In another study, estradiol was the most significant determinant of bone mass only during midpuberty and not before that [61]. A similar result was obtained in boys with constitutional delay of growth and puberty [86]. In this study, the strongest correlation coefficients were found between BMD and serum estradiol levels among hormones, with estradiol being the most potent determinant in pubertal boys. Puberty also interacts with the influence of physical activity on bone. A meta-analysis found gains in BMC due to exercise only in prepubertal subjects [87]. The efficacy of training in terms of bone mineral accrual was substantially affected only by the maturational status of the participants [87].

At the cellular level, it is well known that testosterone and estradiol are very potent stimulators of bone formation, and this knowledge will not be further discussed into detail. Testosterone exerts its effect on bone cells in boys mainly through conversion to estradiol, with its direct actions being more important for the muscle growth and development. A steep rise in testosterone levels in boys and girls was shown to precede the pubertal increase in bone mass [85]. A relationship between polymorphisms of the aromatase gene and BMD was described in young men [88]. The role of androgens for the stimulation of muscle growth, independent of other hormones such as IGF-1, was summarized in reviews by Vanderschueren et al. (2004) and de Oliveira et al. (2012), while the actions of estrogens were described in the review by Clarke et al. (2010) [222389]. The role of androgen aromatization was further emphasized in a randomized trial of the nonaromatizable androgen dihydrotestosterone (DHT) in healthy adult men [90]. In this study, DHT treatment suppressed serum testosterone resulting in bone loss at the lumbar spine as no estradiol was formed [90].

Another hormonal group , well known for its leading role in the development of the muscle–bone unit, comprises growth hormone (GH) and IGF-1. GH can stimulate longitudinal bone growth and increase the effect of physical stress on bone formation [91]. At the cellular level, it stimulates bone formation on endosteal and subperiosteal surfaces. For many years it has been thought that GH exerts its effect on bone mainly through IGF-1 [1]. IGF-1 is involved in direct actions on growth plate chondrocytes and osteogenic cells, building the cortical and trabecular bone. IGF-1 can modulate the production of 1,25(OH)2-vitamin D and the transport system of inorganic phosphate [1]. A recent study has shown, however, that growth hormone can mediate pubertal skeletal development separately from hepatic IGF-1 production [92]. In this study, liver-specific IGF-1-deficient mice were treated with pegvisomant (a GH antagonist), separating the role of GH from IGF-1, whose production by the liver was basically reduced. The amount of cortical tissue formed in the treated mice was substantially lower, showing an independent contribution of GH to the effects of the GH–IGF-1 axis [92]. The strength of the association of IGF-1 versus 25(OH)D with BMC accrual was tested in prepubertal females followed up for a period of up to 9 years [93]. IGF-1 was more strongly associated with BMC accrual than vitamin D at the total body, proximal femur, radius, and lumbar spine. In multistep regression analysis, 25(OH)D did not have a predictive effect on BMC accrual beyond that of IGF-1 [93]. The expression of IGF-1 and its contribution to bone acquisition in mice have been related to the presence of thyroid hormones [94]. Other insulin-like growth factors may also contribute to the acquisition and maintenance of bone mass. In an experimental study, mutations in the insulin-like factor 3 receptor were associated with decreased bone mass in both mice and men (by dual X-ray bone densitometry), as well as with reduced mineralizing surface, bone formation, and osteoclast surface in mice (dynamic histomorphometric and micro-CT analyses) [95].

Adiponectin and leptin are among the other hormones that were studied for their associations with bone growth. The association of adiponectin with bone mass was studied by using DXA and pQCT in the Avon Longitudinal Study of Parents and Children [96]. Adiponectin was found to be inversely related with DXA-derived bone parameters (BMC, bone area) as well as with endosteal relative to periosteal expansion (from pQCT) resulting in increased cortical thickness [96]. The associations of leptin with tibial speed of sound (tSoS) were tested in early and late-pubertal girls [97]. In this study, adiposity and leptin were both negative predictors of tSoS, with leptin being specifically predictive in the postmenarcheal group [97]. Osteosclerostin levels were also studied in relation to bone development during childhood. They were measured in 6–21-year-old girls and boys and related to trabecular and cortical bone microarchitectural parameters using hr-pQCT [98]. Serum sclerostin levels were higher in boys as compared to girls and declined in both sexes following the onset of puberty. The authors found no consistent relationship between sclerostin levels and trabecular bone parameters in either sex, in contrast with an inverse association with cortical vBMD and cortical thickness in girls and a positive one with the cortical porosity index in both girls and boys [98].

In conclusion, bone development during childhood and adolescence is under the influence of so many hormonal factors, whose actions cannot always be differentiated from one another. On the other hand, bone itself may act as an independent endocrine organ, modulating the insulin resistance and secretion, gonadal function, and others [99100]. It should also be noted that the pubertal growth spurt is an event typical for the human species, which cannot be fully reproduced in animal models. Different hypotheses explaining that observation have been proposed, but the evolutionary meaning of these particular time events is not fully understood.

The Use of Bone Age (BA) in Clinical Practice

Skeletal development and maturity have been traditionally studied by the use of bone age (BA) based on hand and wrist radiographs. Two methods are most widely used: the Greulich and Pyle atlas and the Tanner–Whitehouse method [101]. The former is based on comparisons with reference images while the latter—on scores applied to the maturity indicators. At the end of the 1980s, another method was developed—the FELS method [102]. There are many good editions depicting those and other methods for BA assessment [103]. The inter- and intra-observer variability of BA assessment has been a major concern for many years. A recent study in four ethnicities from Los Angeles found a mean standard error of 0.45 years, which is slightly better than the figures reported by older studies [104]. This error might seem significant from a scientific point of view, but is negligible in everyday clinical practice. In recent years, the use of automatic BA assessment has gained more and more followers. For instance, a fully automated computer-based system, which was the first to be introduced in Europe, was the BoneXpert (Visiana, Denmark) [105]. The advantage of this system is the practical lack of variance between readings.

BA is used as the main indicator of skeletal maturity and is routinely compared to chronological age. However, bone mineral accrual and muscle development, height velocity, and others might better correlate with BA than just chronological age. BA is used in a variety of clinical situations [106107]. In precocious puberty BA is a part of the routine work-up and is performed at regular intervals thereafter. It is very useful in decisions regarding treatment. BA is assessed also in cases of premature adrenarche. A recent study found that bone age advancement by 2 or more years is common in children with premature adrenarche and is generally benign [108]. BA is also used in children with unknown chronological age, as well as in delayed puberty. In delayed puberty, BA can help to differentiate between reversible and permanent hypogonadism. Children with skeletal dysplasia remain a difficult population in BA assessment. The hand and wrist radiographs might also be used for other diagnostic purposes, such as in rickets, hypothyroidism, hypochondroplasia, and many others as well as in sports medicine.

Comparisons of different cohorts by the use of BA provided information on secular trends in skeletal maturity. A 35-year difference between the years of birth (1930–1964 versus 1965–2001) resulted in more advanced skeletal maturity (maximum difference of 5 months at age 13 years for girls and 4 months at age 15 years for boys) [109].


Bone development during childhood and adolescence is a very complex process. This complexity mirrors the important evolutionary and biological role of the skeleton. There is a long way to go to the full understanding of all processes implicated in bone growth and development.



Bonjour J-P, Chevalley T, Ferrari S, Rizzoli R. Bone and its characteristics during childhood and adolescence up to peak bone mass. In: Roux C, editor. The living skeleton. Wolters Kluwer Health/Les Laboratories Servier; Rueuil Malmaison, 2007. pp. 1–26.


Maes C, Kronenberg HM. Bone development and remodeling. In: Potts JT Jr, volume editor, Endocrinology: adult and pediatric. 6th ed. The parathyroid gland and bone metabolism. Philadelphia: Elsevier Saunders; 2013. pp. e72–e96.


Provot S, Schipani E, Wu JY, Kronenberg H. Development of the skeleton. In: Marcus R, Feldman D, Dempster DW, Lucky M, Cauley JA, editors. Osteoporosis. 4th ed. Elsevier/Academic Press; Oxford, 2013. pp. 97–126.


Rauch F. The dynamics of bone structure development during pubertal growth. J Musculoskeletal Neuronal Interact. 2012;12(1):1–6.


Emons J, Chagin AS, Sävendahl L, Karperien M, Wit JM. Mechanisms of growth plate maturation and epiphyseal fusion. Horm Res Paediatr. 2011;75(6):383–91.PubMed


Yoon BS, Pogue R, Ovchinnikov DA, Yoshii I, Mishina Y, Behringer RR, et al. BMPs regulate multiple aspects of growth-plate chondrogenesis through opposing actions on FGF pathways. Development. 2006;133(23):4667–78.PubMed


Liu Z, Xu J, Colvin JS, Ornitz DM. Coordination of chondrogenesis and osteogenesis by fibroblast growth factor 18. Genes Dev. 2002;16(7):859–69.PubMedPubMedCentral


Hung IH, Yu K, Lavine KJ, Ornitz DM. FGF-9 regulates early hypertrophic chondrocyte differentiation and skeletal vascularization in the developing stylopod. Dev Biol. 2007;307(2):300–13.PubMedPubMedCentral


Pannier S, Mugniery E, Jonqoy A, Benoist-Lasselin C, Odent T, Jais JP, et al. Delayed bone age due to a dual effect of FGFR3 mutation in achondroplasia. Bone. 2010;47(5):905–15.PubMed


Toydemir RM, Brassington AE, Barak-Toydemir P, Krakowiak PA, Jorde LB, Whitby FG, et al. A novel mutation in FGFR3 causes camptodactyly, tall stature and hearing loss (CATSHL) syndrome. Am J Hum Genet. 2006;79(5):935–41.PubMedPubMedCentral


Day TF, Day BS, Yang Y. Wnt and hedgehog signaling pathways in bone development. J Bone Joint Surg Am. 2008;90 Suppl 1:19–24.PubMed


Emons J, Chagin AS, Malmlöf T, Lekman M, Tivesten A, Ohlsson C, et al. Expression of vascular endothelial growth factor in the growth plate is stimulated by estradiol and increases during pubertal development. J Endocrinol. 2010;205(1):61–8. doi:10.1677/JOE-09-0337.PubMed


Kobayashi T, Chung UI, Schipani E, Starbuck M, Karsenty G, Katagiri T, et al. PTHrp and Indian hedgehog control differentiation of growth plate chondrocytes at multiple steps. Development. 2002;129(12):2977–86.PubMed


Hellemans J, Coucke PJ, Giedion A, De Paepe A, Kramer P, Beemer F, et al. Homozygous mutations in IHH cause acrocapitofemoral dysplasia, an autosomal recessive disorder with cone-shaped epiphyses in hands and hips. Am J Hum Genet. 2003;72(4):1040–6.PubMedPubMedCentral


Risom L, Christoffersen L, Daugaard-Jensen J, Hove HD, Andersen HS, Andresen BS, et al. Identification of six novel PTH1R mutations in families with a history of primary failure of tooth eruption. PLoS One. 2013;8(9), e74601.PubMedPubMedCentral


Bastepe M, Raas-Rothschild A, Silver J, Weissman I, Wientroub S, Jüppner H, et al. A form of Jansen’s metaphyseal chondrodysplasia with limited metabolic and skeletal abnormalities is caused by a novel activating parathyroid hormone (PTH)/PTH-related peptide receptor mutation. J Clin Endocrinol Metab. 2004;89(7):3595–600.PubMed


Idelevich A, Kerschnitzki M, Shahar R, Monsonego-Ornan E. 1,25(OH)2D3 alters growth plate maturation and bone architecture in young rats with normal renal function. PLoS One. 2011;6(6), e20772. doi:10.1371/journal.pone.0020772.PubMedPubMedCentral


Stevens A, Hanson D, Whatmore A, Destenaves B, Chatelain P, Clayton P. Human growth is associated with distinct patterns of gene expression in evolutionarily conserved networks. BMC Genomics. 2013;14:547. doi:10.1186/1471-2164-14-547.PubMedPubMedCentral


Börjesson AE, Windahl SH, Karimian E, Eriksson EE, Lagerquist MK, Engdahl C, et al. The role of estrogen receptor-α and its activation function-1 for growth plate closure in female mice. Am J Physiol Endocrinol Metab. 2012;302(11):E1381–9. doi:10.1152/ajpendo.00646.2011.PubMedPubMedCentral


Börjesson AE, Lagerquist MK, Windahl SH, Ohlsson C. The role of estrogen receptor α in the regulation of bone and growth plate cartilage. Cell Mol Life Sci. 2013;70(21):4023–37. doi:10.1007/s00018-013-1317-1.PubMed


Börjesson AE, Lagerquist MK, Liu C, Shao R, Windahl SH, Karlsson C, et al. The role of estrogen receptor α in growth plate cartilage for longitudinal bone growth. J Bone Miner Res. 2010;25(12):2690–700. doi:10.1002/jbmr.156.PubMed


Vanderschueren D, Vandenput L, Boonen S, Lindberg MK, Bouillon R, Ohlsson C. Androgens and bone. Endocr Rev. 2004;25(3):389–425.PubMed


De Oliveira DH, Fighera TM, Bianchet LC, Kulak CA, Kulak J. Androgens and bone. Minerva Endocrinol. 2012;37(4):305–14.PubMed


Emons J, Chagin AS, Hultenby K, Zhivotovsky B, Wit JM, Karperien M, et al. Epiphyseal fusion in the human growth plate does not involve classical apoptosis. Pediatr Res. 2009;66(6):654–9. doi:10.1203/PDR.0b013e3181beaa8c.PubMed


De la Rica L, Rodríguez-Ubreva J, García M, Islam AB, Urquiza JM, Hernando H, et al. PU.1 target genes undergo Tet2-coupled demethylation and DNMT3b-mediated methylation in monocyte-to-osteoclast differentiation. Genome Biol. 2013;14(9):R99.PubMedPubMedCentral


Kwon OH, Lee CK, Lee YI, Paik SG, Lee HJ. The hematopoietic transcription factor PU.1 regulates RANK gene expression in myeloid progenitors. Biochem Biophys Res Commun. 2005;335(2):437–46.PubMed


Stanley ER, Chitu V. CSF-1 receptor signaling in myeloid cells. Cold Spring Harb Perspect Biol. 2014;6(6):pii: a021857. doi:10.1101/cshperspect.a021857.


Leibbrandt A, Penninger JM. RANK(L) as a key target for controlling bone loss. Adv Exp Med Biol. 2009;647:130–45.PubMed


Schneeweis LA, Willard D, Milla ME. Functional dissection of osteoprotegerin and its interaction with receptor activator of NF-κB ligand. J Adv Exp Med Biol. 2009;647:130–45. doi:10.1007/978-0-387-89520-8_9.


Zhao H, Liu X, Zou H, Dai N, Yao L, Gao Q, et al. Osteoprotegerin induces podosome disassembly in osteoclasts through calcium, ERK, and p38 MAPK signaling pathways. Cytokine. 2015;71(2):199–206. doi:10.1016/j.cyto.2014.10.007.PubMed


Souza PP, Lerner UH. The role of cytokines in inflammatory bone loss. Immunol Invest. 2013;42(7):555–622. doi:10.3109/08820139.2013.822766.PubMed


Wang Y, Qin QH, Kalyanasundaram S. A theoretical model for simulating effect of parathyroid hormone on bone metabolism at cellular level. Mol Cell Biomech. 2009;6(2):101–12.PubMed


Takahashi N, Udagawa N, Suda T. Vitamin D endocrine system and osteoclasts. Bonekey Rep. 2014;3:495. doi:10.1038/bonekey.2013.229.PubMedPubMedCentral


Sims NA, Vrahnas C. Regulation of cortical and trabecular bone mass by communication between osteoblasts, osteocytes and osteoclasts. Arch Biochem Biophys. 2014;561:22–8. doi:10.1016/


Luis Neyro J, Jesús Cancelo M, Palacios S. Inhibition of RANK-L in the pathophysiology of osteoporosis. Clinical evidences of its use. Ginecol Obstet Mex. 2013;81(3):146–57.PubMed


Bonewald LF, Johnson ML. Osteocytes, mechanosensing and Wnt signaling. Bone. 2008;42(4):606–15. doi:10.1016/j.bone.2007.12.224.PubMedPubMedCentral


Galli C, Passeri G, Macaluso GM. Osteocytes and WNT: the mechanical control of bone formation. J Dent Res. 2010;89(4):331–43. doi:10.1177/0022034510363963.PubMed


Kubota T, Michigami T, Ozono K. Wnt signaling in bone. Clin Pediatr Endocrinol. 2010;19(3):49–56. doi:10.1297/cpe.19.49.PubMedPubMedCentral


Baron R, Kneissel M. WNT signaling in bone homeostasis and disease: from human mutations to treatments. Nat Med. 2013;19(2):179–92. doi:10.1038/nm.3074.PubMed


Yavropoulou MP, Xygonakis C, Lolou M, Karadimou F, Yovos JG. The sclerostin story: from human genetics to the development of novel anabolic treatment for osteoporosis. Hormones (Athens). 2014;13(4):323–37. doi:10.14310/horm.2002.1552.


Seeman E. Bone modeling and remodeling. Crit Rev Eukaryot Gene Expr. 2009;19(3):219–33.PubMed


Briot K, Kolta S, Fechtenbaum J, Said-Nahal R, Benhamou CL, Roux C. Increase in vertebral body size in postmenopausal women with osteoporosis. Bone. 2010;47(2):229–34. doi:10.1016/j.bone.2010.03.020.PubMed


Sims NA, Martin TJ. Coupling the activities of bone formation and resorption: a multitude of signals within the basic multicellular unit. Bonekey Rep. 2014;3:481. doi:10.1038/bonekey.2013.215.PubMedPubMedCentral


Feng X, McDonald JM. Disorders of bone remodeling. Annu Rev Pathol. 2011;6:121–45. doi:10.1146/annurev-pathol-011110-130203.PubMedPubMedCentral


Duan Y, Beck TJ, Wang X-F, Seeman E. Structural and biomechanical basis of sexual dimorphism in femoral neck fragility has its origins in growth and aging. J Bone Miner Res. 2003;18(10):1766–74.PubMed


Boyanov M, Popivanov P, Gentchev G. Assessment of forearm volumetric bone mineral density from standard areal densitometry data. J Clin Densitom. 2002;5(4):391–402.PubMed


Boyanov M, Popivanov P, Roux C. Separate assessment of forearm cortical and trabecular bone density from standard densitometry data. Ann Med. 2001;33(7):497–506.PubMed


Ausili E, Rigante D, Salvaggio E, Focarelli B, Rendeli C, Ansuini V, et al. Determinants of bone mineral density, bone mineral content, and body composition in a cohort of healthy children: influence of sex, age, puberty, and physical activity. Rheumatol Int. 2012;32(9):2737–43. doi:10.1007/s00296-011-2059-8.PubMed


Weaver CM, McCabe LD, McCabe GP, Novotny R, Van Loan M, Going S, et al. Bone mineral and predictors of bone mass in white, Hispanic, and Asian early pubertal girls. Calcif Tissue Int. 2007;81(5):352–63.PubMed


Gracia-Marco L, Vicente-Rodríguez G, Valtueña J, Rey-López JP, Díaz Martínez AE, Mesana MI, et al. Bone mass and bone metabolism markers during adolescence: the HELENA Study. Horm Res Paediatr. 2010;74(5):339–50. doi:10.1159/000314965.PubMed


Magarey AM, Boulton TJ, Chatterton BE, Schultz C, Nordin BE, Cockington RA. Bone growth from 11 to 17 years: relationship to growth, gender and changes with pubertal status including timing of menarche. Acta Paediatr. 1999;88(2):139–46.PubMed


Leonard MB, Elmi A, Mostoufi-Moab S, Shults J, Burnham JM, Thayu M, et al. Effects of sex, race, and puberty on cortical bone and the functional muscle bone unit in children, adolescents, and young adults. J Clin Endocrinol Metab. 2010;95(4):1681–9. doi:10.1210/jc.2009-1913.PubMedPubMedCentral


Nishiyama KK, Macdonald HM, Moore SA, Fung T, Boyd SK, McKay HA. Cortical porosity is higher in boys compared with girls at the distal radius and distal tibia during pubertal growth: an HR-pQCT study. J Bone Miner Res. 2012;27(2):273–82. doi:10.1002/jbmr.552.PubMed


Xu L, Nicholson P, Wang Q, Alén M, Cheng S. Bone and muscle development during puberty in girls: a seven-year longitudinal study. J Bone Miner Res. 2009;24(10):1693–8. doi:10.1359/jbmr.090405.PubMed


Naganathan V, Sambrook P. Gender differences in volumetric bone mineral density: a study of opposite-sex twins. Osteoporos Int. 2003;14(7):564–9.PubMed


Bonjour JP, Theintz G, Buchs B, Slosman D, Rizzoli R. Critical years and stages of puberty for spinal and femoral bone mass accumulation during adolescence. J Clin Endocrinol Metab. 1991;73(3):555–63.PubMed


Recker RR, Davies KM, Hinders SM, Heaney RP, Stegman MR, Kimmel DB. Bone gain in young adult women. JAMA. 1992;268(17):2403–8.PubMed


Nguyen TV, Maynard LM, Towne B, Roche AF, Wisemandle W, Li J. Sex differences in bone mass acquisition during growth. The Fels longitudinal study. J Clin Densitom. 2001;4(2):147–57.PubMed


Henry YM, Fatayerji D, Eastell R. Attainment of peak bone mass at the lumbar spine, femoral neck and radius in men and women relative contributions of bone size and volumetric bone mineral density. Osteoporos Int. 2004;15(4):263–73.PubMed


Ivuskans A, Lätt E, Mäestu J, Saar M, Purge P, Maasalu K, et al. Bone mineral density in 11–13-year-old boys: relative importance of the weight status and body composition factors. Rheumatol Int. 2013;33(7):1681–7. doi:10.1007/s00296-012-2612-0.PubMed


Csakvary V, Erhardt E, Vargha P, Oroszlan G, Bodecs T, Torok D, et al. Association of lean and fat body mass, bone biomarkers and gonadal steroids with bone mass during pre- and midpuberty. Horm Res Paediatr. 2012;78(4):203–11. doi:10.1159/000342335.PubMed


Sayers A, Tobias JH. Fat mass exerts a greater effect on cortical bone mass in girls than boys. J Clin Endocrinol Metab. 2010;95(2):699–706. doi:10.1210/jc.2009-1907.PubMed


Boyanov M. Body fat, lean mass and bone density of the spine and forearm in women. Cent Eur J Med. 2014;9(1):121–5. doi:10.2478/s11536-013-0259-1.


Francis SL, Letuchy EM, Levy SM, Janz KF. Sustained effects of physical activity on bone health: Iowa Bone Development Study. Bone. 2014;63:95–100. doi:10.1016/j.bone.2014.03.004.PubMedPubMedCentral


Vaitkeviciute D, Lätt E, Mäestu J, Jürimäe T, Saar M, Purge P, Maasalu K, Jürimäe J. Physical activity and bone mineral accrual in boys with different body mass parameters during puberty: a longitudinal study. PLoS One. 2014;9(10), e107759. doi:10.1371/journal.pone.0107759.PubMedCentral


Ivuškāns A, Mäestu J, Jürimäe T, Lätt E, Purge P, Saar M. Sedentary time has a negative influence on bone mineral parameters in peripubertal boys: a 1-year prospective study. J Bone Miner Metab. 2015;33(1):85–92. doi:10.1007/s00774-013-0556-4.PubMed


Ubago-Guisado E, Gómez-Cabello A, Sánchez-Sánchez J, García-Unanue J, Gallardo L. Influence of different sports on bone mass in growing girls. J Sports Sci. 2015;21:1–9.


Nogueira RC, Weeks BK, Beck BR. Exercise to improve pediatric bone and fat: a systematic review and meta-analysis. Med Sci Sports Exerc. 2014;46(3):610–21. doi:10.1249/MSS.0b013e3182a6ab0d.PubMed


Ishikawa S, Kim Y, Kang M, Morgan DW. Effects of weight-bearing exercise on bone health in girls: a meta-analysis. Sports Med. 2013;43(9):875–92. doi:10.1007/s40279-013-0060-y.PubMed


Schoenau E, Fricke O. Mechanical influences on bone development in children. Eur J Endocrinol. 2008;159 Suppl 1:S27–31. doi:10.1530/EJE-08-0312.PubMed


Kalkwarf KJ, Khoury JC, Lanphear BP. Milk intake during childhood and adolescence, adult bone density and osteoporotic fractures in US women. Am J Clin Nutr. 2003;77(1):257–65.PubMed


Wosje KS, Specker BL. Role of calcium in bone health during childhood. Nutr Rev. 2000;58(9):253–68.PubMed


Bonjour JP, Carrie AL, Ferrari S, Clavien H, Slosman D, Theintz G, et al. Calcium-enriched foods and bone mass growth in prepubertal girls: a randomized, double-blind, placebo-controlled trial. J Clin Invest. 1997;99(6):1287–94.PubMedPubMedCentral


Esterle L, Nguyen M, Walrant-Debray O, Sabatier JP, Garabedian M. Adverse interaction of low-calcium diet and low 25(OH)D levels on lumbar spine mineralization in late-pubertal girls. J Bone Miner Res. 2010;25(11):2392–8. doi:10.1002/jbmr.134.PubMed


Specker B, Binkley T. Randomized trial of physical activity and calcium supplementation on bone mineral content in 3- to 5-year old children. J Bone Miner Res. 2003;18(5):885–92.PubMed


Iuliano-Burns S, Saxon L, Naughton G, Gibbons K, Bass SL. Regional specificity of calcium and exercise during skeletal growth in girls: a randomized controlled trial. J Bone Miner Res. 2003;18(1):156–62.PubMed


Dogan M, Derman O, Oksüz-Kanbur N, Akgül S, Kutluk T. The effects of nutrition and physical activity on bone development in male adolescents. Turk J Pediatr. 2009;51(6):545–50.PubMed


Lappe JM, Watson P, Gilsanz V, Hangartner T, Kalkwarf HJ, Oberfield S, et al. The longitudinal effects of physical activity and dietary calcium on bone mass accrual across stages of pubertal development. J Bone Miner Res. 2015;30(1):156–64. doi:10.1002/jbmr.2319.PubMedPubMedCentral


Chevalley T, Bonjour JP, van Rietbergen B, Ferrari S, Rizzoli R. Tracking of environmental determinants of bone structure and strength development in healthy boys: an eight-year follow up study on the positive interaction between physical activity and protein intake from prepuberty to mid-late adolescence. J Bone Miner Res. 2014;29(10):2182–92. doi:10.1002/jbmr.2247.PubMed


Lucas R, Fraga S, Ramos E, Barros H. Early initiation of smoking and alcohol drinking as a predictor of lower forearm bone mineral density in late adolescence: a cohort study in girls. PLoS One. 2012;7(10), e46940. doi:10.1371/journal.pone.0046940.PubMedPubMedCentral


Zemel B. Bone mineral accretion and its relationship to growth, sexual maturation and body composition during childhood and adolescence. World Rev Nutr Diet. 2013;106:39–45. doi:10.1159/000342601.PubMed


Gilsanz V, Chalfant J, Kalkwarf H, Zemel B, Lappe J, Oberfield S, et al. Age at onset of puberty predicts bone mass in young adulthood. J Pediatr. 2011;158(1):100–105.e2. doi:10.1016/j.jpeds.2010.06.054.PubMed


Schoenau E. Bone mass increase in puberty: what makes it happen? Horm Res. 2006;65 Suppl 2:2–10.PubMed


Pérez-López FR, Chedraui P, Cuadros-López JL. Bone mass gain during puberty and adolescence: deconstructing gender characteristics. Curr Med Chem. 2010;17(5):453–66.PubMed


Yilmaz D, Ersoy B, Bilgin E, Gümüşer G, Onur E, Pinar ED. Bone mineral density in girls and boys at different pubertal stages: relation with gonadal steroids, bone formation markers, and growth parameters. J Bone Miner Metab. 2005;23(6):476–82.PubMed


Doneray H, Orbak Z. Association between anthropometric hormonal measurements and bone mineral density in puberty and constitutional delay of growth and puberty. West Indian Med J. 2010;59(2):125–30.PubMed


Behringer M, Gruetzner S, McCourt M, Mester J. Effects of weight-bearing activities on bone mineral content and density in children and adolescents: a meta-analysis. J Bone Miner Res. 2014;29(2):467–78. doi:10.1002/jbmr.2036.PubMed


Riancho JA. Polymorphisms in the CYP19 gene that influence bone mineral density. Pharmacogenomics. 2007;8(4):339–52.PubMed


Clarke BL, Khosla S. Female reproductive system and bone. Arch Biochem Biophys. 2010;503(1):118–28.PubMedPubMedCentral


Sartorius GA, Ly LP, Handelsman DJ. Male sexual function can be maintained without aromatization: randomized placebo-controlled trial of dihydrotestosterone (DHT) in healthy, older men for 24 months. J Sex Med. 2014;11(10):2562–70.PubMed


Forwood MR, Li L, Kelly WL, Bennet MB. Growth hormone is permissive for skeletal adaptation to mechanical loading. J Bone Miner Res. 2001;16(12):2284–90.PubMed


Courtland H-W, Sun H, Beth-On M, Wu Y, Elis S, Rosen CJ, et al. Growth hormone mediates pubertal skeletal development independent of hepatic IGF-1 production. J Bone Miner Res. 2011;26(4):761–8. doi:10.1002/jbmr.265.PubMed


Breen ME, Laing EM, Hall DB, Hausman DB, Taylor RG, Isales CM, Ding KH, Pollock NK, Hamrick MW, Baile CA, Lewis RD. 25-hydroxyvitamin D, insulin-like growth factor-I, and bone mineral accrual during growth. J Clin Endocrinol Metab. 2011;96(1):E89–98. doi:10.1210/jc.2010-0595.PubMed


Xing W, Govoni KE, Donahue LR, Kesavan C, Wergedal J, Long C, Bassett JH, Gogakos A, Wojcicka A, Williams GR, Mohan S. Genetic evidence that thyroid hormone is indispensable for prepubertal insulin-like growth factor-I expression and bone acquisition in mice. J Bone Miner Res. 2012;27(5):1067–79. doi:10.1002/jbmr.1551.PubMed


Ferlin A, Pepe A, Gianesello L, Garolla A, Feng S, Giannini S, et al. Mutations in the insulin-like factor 3 receptor are associated with osteoporosis. J Bone Miner Res. 2008;23(5):683–93.PubMedPubMedCentral


Sayers A, Timpson NJ, Sattar N, Deanfield J, Hingorani AD, Davey-Smith G, et al. Adiponectin and its association with bone mass accrual in childhood. J Bone Miner Res. 2010;25(10):2212–20. doi:10.1002/jbmr.116.PubMed


Klentrou P, Ludwa IA, Falk B. Factors associated with bone turnover and speed of sound in early and late-pubertal females. Appl Physiol Nutr Metab. 2011;36(5):707–14. doi:10.1139/h11-085.PubMed


Kirmani S, Amin S, McCready LK, Atkinson EJ, Melton III LJ, Müller R, et al. Sclerostin levels during growth in children. Osteoporos Int. 2012;23(3):1123–30. doi:10.1007/s00198-011-1669-z.PubMed


Guntur AR, Rosen CJ. Bone as an endocrine organ. Endocr Pract. 2012;18(5):758–62.PubMedPubMedCentral


Zofkova I. Bone tissue as a systemic endocrine regulator. Physiol Res. 2014;3 [Epub ahead of print].


Martin DD, Wit JM, Hochberg Z, Sävendahl L, van Rijn RR, Fricke O, Cameron N, Caliebe J, Hertel T, Kiepe D, Albertsson-Wikland K, Thodberg HH, Binder G, Ranke MB. The use of bone age in clinical practice: part 1. Horm Res Paediatr. 2011;76(1):1–9.PubMed


De Sanctis V, Di Maio S, Soliman AT, Raiola G, Elalaily R, Millimaggi G. Hand X-ray in pediatric endocrinology: Skeletal age assessment and beyond. Indian J Endocrinol Metab. 2014;18 Suppl 1:S63–71.PubMedPubMedCentral


Gilsanz V, Ratib O. Hand bone age. A digital atlas of skeletal maturity. 2nd ed. Berlin: Springer; 2012.


Thodberg HH, Sävendahl L. Validation and reference values of automated bone age determination for four ethnicities. Acad Radiol. 2010;17(11):1425–32. doi:10.1016/j.acra.2010.06.007.PubMed


Thodberg HH, Kreiborg S, Juul A, Pedersen KD. The BoneXpert method for automated determination of skeletal maturity. IEEE Trans Med Imaging. 2009;28(1):52–66.PubMed


Martin DD, Wit JM, Hochberg Z, van Rijn RR, Fricke O, Werther G, Cameron N, Hertel T, Wudy SA, Butler G, Thodberg HH, Binder G, Ranke MB. The use of bone age in clinical practice: part 2. Horm Res Paediatr. 2011;76(1):10–6.PubMed


Lazar L, Phillip M. Pubertal disorders and bone maturation. Endocrinol Metab Clin North Am. 2012;41(4):805–25.PubMed


DeSalvo DJ, Mehra R, Vaidyanathan P, Kaplowitz PB. In children with premature adrenarche, bone age advancement by 2 or more years is common and generally benign. J Pediatr Endocrinol Metab. 2013;26(3-4):215–21. doi:10.1515/jpem-2012-0283.PubMed


Duren DL, Nahhas RW, Sherwood RJ. Do secular trends in skeletal maturity occur equally in both sexes? Clin Orthop Relat Res. 2015;473:2559–67.PubMedPubMedCentral