Bone Age

Anatomy and Physiology

1. Visualization by hand & wrist plain film radiographs: Standard posteroanterior (PA) view of the hand and wrist; Effective radiation received during each exposure is between 0.0001-0.1 mSV.[1] Typically, left-hand radiographs are utilized (as most individuals are right dominant and the left hand has less chance of being deformed)

Techniques:

A. Greulich and Pyle (GP) atlas [2][3][4]: Based on "The Radiographic Atlas of Skeletal Development of the Hand and Wrist" by Dr. William Walter Greulich and Dr. Sarah Idell Pyle (1959). Includes reference standards of the left hand and left wrist until 18 years for females and 19 years for males. Advantage: Simpler and faster method; good reliability for Australian and Middle Eastern ethnicity. Disadvantage: Less reliable in Asians.

B. Tanner Whitehouse (TW2) Method [5]: Based on the level of maturity for 20 selected regions of interest (ROI) in specific bones of hand and wrist in each age population. The development level for each ROI is categorized into specific stages, and given a numerical score for each individual bone. A total maturity score is calculated by summing up all these scores, which then correlates with sex-specific bone age. Advantage: More accurate and more reproducible. Disadvantage: Comparatively more complex and more time-consuming.

C. Gilsanz and Ratibin (GR) Atlas [6][7]: Digital atlas prepared by Vincent Gilsanz and Osman Ratibin (2005). Includes age- and sex-specific artificial images of skeletal maturity after thoroughly analyzing size, shape, morphology, and density of ossification centers in healthy children (spaced at 6-monthly intervals between 2 and 6 years, and yearly intervals between 7 and 17 years). Advantage: More precise and better quality images, as compared with GP atlas. Disadvantage: This shows fairly similar results in previous studies. However, GR atlas has more outliers.  

D. Automated skeletal bone age assessment [8][9]: Digital radiographs of hand and wrist Xray - Undergoes several stages of processing:

• Stage 1: Pre-processing: Image normalized to grayscale, background removed and re-oriented
• Stage 2: Segmentation: Desired parts of image separated from the background
• Stage 3: Analysis: Selected regions of interest analyzed for bone age by TW2 method or GP atlas

Recently developed software computes bone age by both GP and TW2 methods, which has demonstrated validity across different ethnicities. 

E. Capito-hamate planimetry [Choi et al. (2018)] [10]: Defined as the measurement of the sum of areas of the capitate and hamate. Can be assessed using plain radiographs.

2. Visualization by ultrasound (US) [11][12][13]: Specialized ultrasound device involving two transducers: one transmitted (ultrasonic waves of 750kHz Frequency) and one receiver. Includes 11 cycles directed at the epiphysis of distal radius and ulna and skeletal age is determined, based on demographics of subject and ultrasound results. Disadvantage: It still needs to be evaluated in multiethnic populations with a large sample size.

3. Visualization by Magnetic Resonance Imaging (MRI) [14]: Hand and wrist MRI has been evaluated as an option for bone age estimation (GP and TW2 methods) (Urschler et al. 2016); nevertheless, this technique still needs further validation. 

4.  Visualization of elbow ossification [15]: Sauvegrain et al. (1962) developed a full 27-point scoring system for bone age assessment, including the assessment of four growth centers (lateral condyle and epicondyle, trochlea, olecranon apophysis, and proximal radial epiphysis) on AP and lateral radiographs of the left elbow. This assessment modality involves relatively greater radiation exposure. The order of appearance and fusion of different ossification centers around the elbow and their correlation with chronological age has been described [16]: capitellum (0-1 year; 10-15 years), radial head (2-6 years; 12-16 years), medial epicondyle (2-8 years; 13 years), trochlea (5-11 years; 10-18 years), olecranon (6-11 years; 13-16 years) and lateral epicondyle (8-13 years; 12-16 years). Appearance and fusion for all these ossification centers (except capitellum and radial head) have been demonstrated to be earlier in girls than boys.

5. Visualization of humeral head ossification [17]: Li et al. (2018) had recently described the 5-staged humeral head ossification classification system:

• Stage 1 includes an incompletely ossified lateral epiphysis
• Stage 2 demonstrates increased ossification of the lateral epiphysis, with a lateral curved margin
• Stages 3 to 5 demonstrate collinearity between lateral margin of epiphysis and metaphysis; in stage 3, the lateral half of the physis is open without obvious fusion, in stage 4, the lateral half of physis is partly fused, and in stage 5, the lateral half of physis is completely fused

It has been demonstrated that peak height velocity (PHV) correlates between stages 2 and 3.

6. Visualization of the clavicle [18][19][20]: A secondary ossification center develops at the medial end of the clavicle during adolescence and undergoes fusion at 22 years. Best bone to image for bone age assessment between 18 and 22 years of age. Conventional radiographs - not much use to assess; Conventional or Spiral CT - Concerns regarding radiation exposure; and MRI (3-Tesla versus 1.5 Tesla) - Can be a radiation-free option. However, further research to develop specific protocols needed.

7. Visualization of iliac bone [21][22]: 

A. By pelvic radiographs: Risser sign - Based on the ossification of iliac crest apophysis. However, not a reliable or commonly used technique.

B. Non-ionizing Imaging: Ultrasonography as a technique to assess bone age using iliac crest apophysis ossification - has also been considered. However, this technique still requires validation.

8. Visualization of the femoral head [7]: Bone age can be calculated by assessing the depth of epiphyseal cartilage of the femoral head, in contrast to visualizing bony epiphyseal femur. As the ossification process occurs, most of the cartilage undergoes replacement by bone and hyaline articular cartilage. Ultrasound assessment has been studied to evaluate femoral head ossification; however, the evidence is still unequivocal. 

9. Modified Oxford Score [23][24]: Described by Stasikelis et al. assessed bone age on the basis of three consecutive stages of maturation of triradiate cartilage (TRC), femoral epiphysis, greater trochanter, and lesser trochanter. The total score ranges from 16 to 26. There is a solid body of literature regarding the role of this technique in predicting the contralateral slip in patients with SCFE. Modified Oxford score of 16 to 18 corresponds to increased contralateral slips in SCFE, and 19 to 21 corresponds to PHV.

10. Calcaneal apophyseal ossification [25]: Nicholson et al. (2016) described a 5-staged bone age assessment based on the calcaneal apophysis. Stage 0 - No ossification of the apophysis

• Stage 1 - Apophysis covers less than 50% of the metaphysis
• Stage 2 - Apophysis covers more than 50% of the metaphysis
• Stage 3 - Apophysis has extended fully over the plantar surface and continues to extend over the dorsal surface without evidence of fusion
• Stage 4 - Fusion of the apophysis to metaphysis begins
• Stage 5 - Fusion is complete

Calcaneal stages 0 to 2 corresponded to high chances of the contralateral slip in SCFE, and stage 3  corresponds closely to PHV

11. Visualization of cervical maturation [26]: Schelgl et al. (2017) described the role of EOS 2D/3D imaging in the assessment of bone age, based on cervical vertebral body morphologies (Hassel–Farman Bone Age Stages - initiation, acceleration, transition, deceleration, maturation and completion stages).

12. Visualization of Dental Maturity: Predominantly studies for forensic purposes. Mineralization of teeth is much less affected by nutritional or endocrine disorders than the skeletal system. 

Techniques:

A. Atlas method [27]: Based on atlas containing standard age-matched orthopantomographic images, developed by Scour et al. (1944), This was further modified by Moorrees et al. and Anderson et al., where they defined dentition stages of all teeth.

B. Scoring method [27][7]: Numerical method devised by Demirjian et al. (1973), involving maturity score of each tooth, based on the level of dentition. A total maturity score is calculated by adding individual maturity scores. It has been demonstrated to show a good correlation with the GP method of bone age estimation.

13. Chondrocyte physiology: The thyroid hormone regulates the growth and maturation of bones by regulating cartilage proliferation, differentiation, and osteogenesis. It facilitates growth hormone action. Cartilage is made up of cells (chondroblasts and chondrocytes) and extracellular matrix (10% aggrecan, 75% water, collagen, and other constituents). Chondroblasts initially secrete the extracellular matrix, inside which they become trapped and mature into chondrocytes. The outermost layer of cartilage is a layer of dense irregular connective tissue called perichondrium, which is constituted by an inner layer of chondroblasts and an outer layer of fibroblasts. In growing cartilage, chondrocytes divide and develop into nests of 2 to 4 cells. These cells are situated in matrix-enclosed compartments called lacunae. The size of its lacuna is a good criterion in deciding whether a chondrocyte is young or old. The larger the lacuna, the older the chondrocyte and vice versa. Active, young chondrocytes are secretory and contain basophilic cytoplasm with the rich, rough endoplasmic reticulum (RER). Older chondrocytes contain fat droplets.

Indications

Bone age assessment is a requirement in the following:

• Endocrine/metabolic/pediatric conditions [28][7][29]: 
  • Diagnosis and management of endocrine disorders 
  • Evaluation of metabolic growth disorders (tall/short stature)
  • Deceleration of maturity in various syndromic disorders; and
  • Assessment of treatment response in various developmental disorders

• Bone malformations and skeletal dysplasias for medicolegal purposes in children and adults: estimation of chronological age when accurate birth records are not available, immigration and other lawsuits, competitive sports, etc.[30][31]
• Computing the final height of normal, healthy children
• Management of orthopedic conditions in children and adolescents
  • Slipped capital femoral epiphysis (SCFE)[24]
  • Management of early-onset and adolescent idiopathic scoliosis[32]
  • Management of limb length discrepancies or limb deformities or physeal arrest[33] - To assess the predicted lower extremity LLD and timing of epiphysiodesis
• Anthropology
• Orthodontics and jaw orthopedics

Equipment

Traditional imaging modality for bone age assessment:

• Plain radiographs; Disadvantage: radiation exposure, inadequate imaging for clavicle

Alternate modalities:

• Modalities without radiation exposure risk: ultrasound (US), magnetic resonance imaging (MRI)
• Modalities with 3-D imaging: computerized tomography (CT)

Technique or Treatment

 1. Visualization by hand & wrist plain film radiographs: Standard posteroanterior (PA) view of the hand and wrist; Effective radiation received during each exposure is between 0.0001-0.1 mSV.[1] Typically, left-hand radiographs are utilized (as most individuals are right dominant and the left hand has less chance of being deformed)

Techniques:

A. Greulich and Pyle (GP) atlas [2][3][4]: Based on "The Radiographic Atlas of Skeletal Development of the Hand and Wrist" by Dr. William Walter Greulich and Dr. Sarah Idell Pyle (1959). Includes reference standards of the left hand and left wrist until 18 years for females and 19 years for males. Advantage: Simpler and faster method; good reliability for Australian and Middle Eastern ethnicity. Disadvantage: Less reliable in Asians.

B. Tanner Whitehouse (TW2) Method [5]: Based on the level of maturity for 20 selected regions of interest (ROI) in specific bones of hand and wrist in each age population. The development level for each ROI is categorized into specific stages, and given a numerical score for each individual bone. A total maturity score is calculated by summing up all these scores, which then correlates with sex-specific bone age. Advantage: More accurate and more reproducible. Disadvantage: Comparatively more complex and more time-consuming.

C. Gilsanz and Ratibin (GR) Atlas [6][7]: Digital atlas prepared by Vincent Gilsanz and Osman Ratibin (2005). Includes age- and sex-specific artificial images of skeletal maturity after thoroughly analyzing size, shape, morphology, and density of ossification centers in healthy children (spaced at 6-monthly intervals between 2 and 6 years, and yearly intervals between 7 and 17 years). Advantage: More precise and better quality images, as compared with GP atlas. Disadvantage: This shows fairly similar results in previous studies. However, GR atlas has more outliers.  

D. Automated skeletal bone age assessment [8][9]: Digital radiographs of hand and wrist Xray - Undergoes several stages of processing:

• Stage 1: Pre-processing: Image normalized to grayscale, background removed and re-oriented
• Stage 2: Segmentation: Desired parts of image separated from the background
• Stage 3: Analysis: Selected regions of interest analyzed for bone age by TW2 method or GP atlas

Recently developed software computes bone age by both GP and TW2 methods, which has demonstrated validity across different ethnicities. 

E. Capito-hamate planimetry [Choi et al. (2018)] [10]: Defined as the measurement of the sum of areas of the capitate and hamate. Can be assessed using plain radiographs.

2. Visualization by ultrasound (US) [11][12][13]: Specialized ultrasound device involving two transducers: one transmitted (ultrasonic waves of 750kHz Frequency) and one receiver. Includes 11 cycles directed at the epiphysis of distal radius and ulna and skeletal age is determined, based on demographics of subject and ultrasound results. Disadvantage: It still needs to be evaluated in multiethnic populations with a large sample size.

3. Visualization by Magnetic Resonance Imaging (MRI) [14]: Hand and wrist MRI has been evaluated as an option for bone age estimation (GP and TW2 methods) (Urschler et al. 2016); nevertheless, this technique still needs further validation. 

4.  Visualization of elbow ossification [15]: Sauvegrain et al. (1962) developed a full 27-point scoring system for bone age assessment, including the assessment of four growth centers (lateral condyle and epicondyle, trochlea, olecranon apophysis, and proximal radial epiphysis) on AP and lateral radiographs of the left elbow. This assessment modality involves relatively greater radiation exposure. The order of appearance and fusion of different ossification centers around the elbow and their correlation with chronological age has been described [16]: capitellum (0-1 year; 10-15 years), radial head (2-6 years; 12-16 years), medial epicondyle (2-8 years; 13 years), trochlea (5-11 years; 10-18 years), olecranon (6-11 years; 13-16 years) and lateral epicondyle (8-13 years; 12-16 years). Appearance and fusion for all these ossification centers (except capitellum and radial head) have been demonstrated to be earlier in girls than boys.

5. Visualization of humeral head ossification [17]: Li et al. (2018) had recently described the 5-staged humeral head ossification classification system:

• Stage 1 includes an incompletely ossified lateral epiphysis
• Stage 2 demonstrates increased ossification of the lateral epiphysis, with a lateral curved margin
• Stages 3 to 5 demonstrate collinearity between lateral margin of epiphysis and metaphysis; in stage 3, the lateral half of the physis is open without obvious fusion, in stage 4, the lateral half of physis is partly fused, and in stage 5, the lateral half of physis is completely fused

It has been demonstrated that peak height velocity (PHV) correlates between stages 2 and 3.

6. Visualization of the clavicle [18][19][20]: A secondary ossification center develops at the medial end of the clavicle during adolescence and undergoes fusion at 22 years. Best bone to image for bone age assessment between 18 and 22 years of age. Conventional radiographs - not much use to assess; Conventional or Spiral CT - Concerns regarding radiation exposure; and MRI (3-Tesla versus 1.5 Tesla) - Can be a radiation-free option. However, further research to develop specific protocols needed.

7. Visualization of iliac bone [21][22]: 

A. By pelvic radiographs: Risser sign - Based on the ossification of iliac crest apophysis. However, not a reliable or commonly used technique.

B. Non-ionizing Imaging: Ultrasonography as a technique to assess bone age using iliac crest apophysis ossification - has also been considered. However, this technique still requires validation.

8. Visualization of the femoral head [7]: Bone age can be calculated by assessing the depth of epiphyseal cartilage of the femoral head, in contrast to visualizing bony epiphyseal femur. As the ossification process occurs, most of the cartilage undergoes replacement by bone and hyaline articular cartilage. Ultrasound assessment has been studied to evaluate femoral head ossification; however, the evidence is still unequivocal. 

9. Modified Oxford Score [23][24]: Described by Stasikelis et al. assessed bone age on the basis of three consecutive stages of maturation of triradiate cartilage (TRC), femoral epiphysis, greater trochanter, and lesser trochanter. The total score ranges from 16 to 26. There is a solid body of literature regarding the role of this technique in predicting the contralateral slip in patients with SCFE. Modified Oxford score of 16 to 18 corresponds to increased contralateral slips in SCFE, and 19 to 21 corresponds to PHV.

10. Calcaneal apophyseal ossification [25]: Nicholson et al. (2016) described a 5-staged bone age assessment based on the calcaneal apophysis. Stage 0 - No ossification of the apophysis

• Stage 1 - Apophysis covers less than 50% of the metaphysis
• Stage 2 - Apophysis covers more than 50% of the metaphysis
• Stage 3 - Apophysis has extended fully over the plantar surface and continues to extend over the dorsal surface without evidence of fusion
• Stage 4 - Fusion of the apophysis to metaphysis begins
• Stage 5 - Fusion is complete

Calcaneal stages 0 to 2 corresponded to high chances of the contralateral slip in SCFE, and stage 3  corresponds closely to PHV

11. Visualization of cervical maturation [26]: Schelgl et al. (2017) described the role of EOS 2D/3D imaging in the assessment of bone age, based on cervical vertebral body morphologies (Hassel–Farman Bone Age Stages - initiation, acceleration, transition, deceleration, maturation and completion stages).

12. Visualization of Dental Maturity: Predominantly studies for forensic purposes. Mineralization of teeth is much less affected by nutritional or endocrine disorders than the skeletal system. 

Techniques:

A. Atlas method [27]: Based on atlas containing standard age-matched orthopantomographic images, developed by Scour et al. (1944), This was further modified by Moorrees et al. and Anderson et al., where they defined dentition stages of all teeth.

B. Scoring method [27][7]: Numerical method devised by Demirjian et al. (1973), involving maturity score of each tooth, based on the level of dentition. A total maturity score is calculated by adding individual maturity scores. It has been demonstrated to show a good correlation with the GP method of bone age estimation.

Clinical Significance

Bone age assessment: Best imaging - based on age:

• Children less than 18 years:  Left hand and wrist radiograph
• 18-22 years: Medial end of the clavicle (CT versus MRI); sexual, dental, and hand maturity indicators are of not much use in this age group

Specific clinical significance of skeletal indicators in orthopedic surgery:

Scoliosis [34][32]: Early-onset scoliosis (EOS) and adolescent idiopathic scoliosis (AIS) management involve the understanding of a child's remaining growth and skeletal maturity. Pediatric spinal growth classifies into 3 phases:

• Phase 1 - 0 to 5 years
• Phase 2 - 5 to 10 years
• Phase 3 is beyond 10 years

The most rapid growth happens during phases 1 and 3. Curve acceleration phase (CAP) is a time of rapid change in curve magnitude, and patterns of curve progression have been described as low, moderate, and rapid. Sanders et al.[[34] demonstrated a high correlation between the TW2 method of skeletal maturation and CAP. Other maturity indicators related to CAP include Risser stage 0, triradiate closure, PHV, and stage-2 tanner sexual characteristics staging. Sanders et al.[35] developed a simplified bone age assessment and developed an 8-stage classification system for phalangeal maturity. Phalangeal capping (seen in stages 3 and 4) corresponded well with PHV. Although there is a potential role for cervical vertebral body maturation parameters in scoliosis management, the exact correlation with curve progression is still not clear. Modified Oxford score of 19 to 21 correlated well with PHV. Calcaneal ossification stage 3 and humeral ossification stage 2 to 3 also correlate well with PHV.

Risk factors for curve progression (at presentation) include a curve of magnitude greater than 25° before skeletal maturity will continue to progress after skeletal maturity, greater than 50° thoracic curve will progress at the rate of 1 to 2° per year, and greater than 40° lumbar curve will progress at the rate of 1 to 2° per year. Other risk factors for curve progression are remaining skeletal growth, younger age (less than 12 years) at presentation, Tanner stage (less than 3 for females), Risser stage (0-1), and open triradiate cartilage. Peak growth velocity is considered the best predictor of curve progression. In females, growth spurt occurs just before menarche and before Risser 1. Also, if the curve is greater than 30° before peak height velocity, there is a strong likelihood of the need for surgery. The thoracic curve is more likely to progress than lumber. Double major curves are more likely to progress than single curves.

Limb length discrepancies [33]: Assessment of bone age is required to evaluate predicted lower extremity LLD and timing of epiphysiodesis in limb length discrepancies. Planning of surgical management of lower limb deformity (varus and valgus malalignment) also requires a precise assessment of skeletal age. Hand radiographs (GP method) can help us to predict the skeletal age and growth remaining.

SCFE: The modified Oxford bone age score (16-18) has been described as the best predictor of contralateral slipped capital femoral epiphysis in patients with unilateral SCFE.[36] Calcaneal apophysis (stages 0-2) has also been recently described to demonstrate a good correlation to contralateral slip.

Enhancing Healthcare Team Outcomes

Bone age assessment can be of utmost importance in various complicated situations involving medical, surgical, forensic, and legal issues. Apart from the ability to assess the chronological age of a person, the evaluation of growth remaining in a person can be extremely beneficial to a clinician in different circumstances, as previously discussed. Although the role of hand X-ray in skeletal maturity estimation is well-known, evaluation of multiple other radiographs can be more beneficial in certain circumstances [e.g., Oxford hip score and calcaneal apophysis in slipped capital femoral epiphysis (SCFE)]. The evidence on this subject is still not extensive, and future research can pave the way for a better understanding of each of these skeletal assessment modalities. Future research can also help us with developing newer modalities, not involving significant radiation exposure for regular assessment of bone age in young children and adolescents. Also, more automated software to assess bone age can revolutionize this field substantially. 

The radiologist plays a vital role in the assessment of bone age. However, patient management also involves an extensive inter-professional team of specialists, including family physicians, endocrinologists, orthopedists, or forensic experts. Nurses are also a vital part of the treating team as they can help coordinate the long-term management of these children. 

 Level of Evidence - The majority of the studies - are based on Level V (or level IV) evidence.