Survey of pediatric radiologists suggest need for better bone age assessment protocols

Bone age assessment is important for diagnosing and treating endocrine disorders in children and conditions associated with delayed or accelerated skeletal development. A variety of assessment techniques exist and are used, but these have been publicly criticized as being inappropriate and inexact for the population of the 21st century.

Members of the Society of Pediatric Radiology (SPR) were surveyed to determine the techniques they use and to obtain their opinions about their confidence in the established methods. Researchers at Boston Children’s Hospital reported survey findings in an article recently published in Pediatric Radiology.

Bone age represents the degree of secondary ossification in long and short bones. Radiologic bone age assessment can identify an overproduction of growth hormone, thyroid hormone, estrogen, and androgen. Radiologic methods rely on visual inspection of developing ossification centers. These methods include an atlas approach matching templates and single bone approaches that assign bone ages to various ossification centers and combine them to yield a maturation score.

Lead author radiologist Micheál A. Breen, MD, and colleagues in the Department of Radiology surveyed all active members of the SPR. They received a response rate of 47.1%. Nearly two-thirds of the 441 respondents had more than 10 years’ experience as pediatric radiologists. More than half (56.8%) performed bone age assessment from one to 10 times a week.

The most method of assessment reported for all aged children was use of the 1959 edition of Greulich and Pyle’s atlas.¹ The atlas contains reference images of male and female standards of the left wrist and hand from birth to age 18 for girls and 19 for boys. An explanation is provided with each standard image with respect to gradual age-related changes observed in the bone structure. Radiologists calculate bone age by comparing the left wrist radiograph of a patient with the nearest matching reference radiograph provided in the atlas. This method is considered simpler and faster than other radiograph based methods. It was used for infants by 69.8% of respondents, for toddlers aged 1-3 by 85.9%, and for older children by 97.4%.

Other assessments used for infants included the hemiskeleton epiphyseal counting method (26.8%) and the Pyle and Hoerr knee method (14.4%). For toddlers, 18.7% used hemiskeletal epiphyseal counting methods and 7.1% used the knee method.

All but 7% of responding pediatric radiologists expressed confidence in the atlas comparison method for children over the age of three. However, 22% reported they were either somewhat unconfident or very unconfident in using this technique for toddlers, and 34% were equally unconfident for infants. Some radiologists voluntarily added comments to the survey stating that they found it difficult to justify the additional time of acquisition/interpretation and the radiation dose associated with the hemiskeleton techniques. The authors also speculated that “because of their familiarity with Greulich and Pyle in older children, many radiologists simply do not actively consider an alternative method when asked to assess bone age in an infant or young child.” They recommended that development of an accurate, precise, and reproducible method, especially for assessment of infants, be undertaken.

“The survey was inspired by work my colleagues and I performed examining the utility of measuring the length of the fibula on AP radiographs as an alternative method for bone age assessment in younger patients. Our data showed that this novel method is more accurate than the standard methods in infants and younger children; the technique is also simple, objective and takes little time. This work has been published in Pediatric Radiology.² We strongly believe that once our findings are validated by other investigators, it will become  the primary technique for evaluating bone age in infants,” Dr. Breen told Applied Radiology.

One automated bone age determination method (BoneXpert, Søllerødvej, Denmark) has been developed and validated in Caucasian children with various growth disorders. A clinical trial to establish a bone age database using an ultrasound system (SonicBone, Lezion, Israel) designed to measure wrist bones, Phalanx III, and metacarpal bones, was started in June 2013 and is still recruiting patients.³ Another clinical trial is being conducted by researchers at the Sheffield Children’s NHS Foundation Trust in the United Kingdom.⁴ This prospective study is assessing the use of COMBAT1, a computerized bone age tool that utilizes digital bone densitometry (DXA) scans.

REFERENCES

1. Greulich WW, Pyle SI. Radiograph atlas of skeletal development of the hand and wrist. 2nd ed. California: Stanford University Press; 1959.
2. Tsai A, Stamoulis C, Bixby SD, et al. Infant bone age estimation based on fibular shaft length: model development and clinical validation. Pediatr Radiol. 2016 46;3:342-356.
3. https://clinicaltrials.gov/ct2/show/NCT01980693?term=%22bone+age%22&rank=1
4. https://clinicaltrials.gov/ct2/show/NCT02617901?term=%22bone+age%22&rank=2
5. Breen MA, Tsai A, Stamm, et al. Bone age assessment practices in infants and older children among Society for Pediatric Radiology Members. Pediatr Radiol. Epub ahead of print on May 12, 2016. DOI 10.1007/s00247-016-3618-7.
6. Satoh M. Bone age: assessment methods and clinical applications. Clin Pediatr Endocrinol. 2015 24;4: 143-152.