Study Examines Accuracy and Variability of Skeletal Maturity Measurement Systems

A study presented at the AAOS 2024 Annual Meeting in San Francisco examined various skeletal maturity measurement system, finding that systems with more stages and parameters, such as the Modified Fels system, are more precise. However, the differences between systems are minimal, making the choice of system less critical for clinical use. For AAOS Now, Selina C. Poon, MD, MPH, FAAOS, sat down with study co-author Kallie Chen, MD, an orthopaedic surgery resident at University Hospital, Case Western in Cleveland. Dr. Chen also completed the Allen Fellowship, working with her research mentor Raymond W. Liu, MD, FAAOS, at the same institution. Dr. Chen discussed the study’s findings as well as future work, which will focus on automating these systems to improve reliability and efficiency of estimating skeletal maturity for pediatric orthopaedic surgeons.

Dr. Poon: Please tell us a little bit about the project.
Dr. Chen: For this project, we investigated several different skeletal maturity systems that Dr. Liu and his team have worked hard over the years to develop, along with other research teams, to see how well each system lines up with the others. Because, for example, a child who has a shoulder and a knee X-ray that are taken on the same day should, in theory, have the same skeletal age estimate. We wanted to examine the correlation between the different skeletal maturity systems.

Was there a specific event that sparked the research question?
Dr. Liu’s team had previously presented on the Modified Fels knee skeletal maturity system and the app that was developed. During the presentation, someone in the audience asked a question about the app and stated that they noticed the ages of the various skeletal maturity systems do not line up. That question was what actually sparked this project.

What was the most important finding that this audience should take away?
For this project, our main finding was that systems that tend to have more stages and more parameters are more precise [at measuring skeletal maturity]. For example, the Modified Fels knee skeletal maturity system, [which] has seven parameters calculated from X-rays, tends to be more precise than systems that have a single stage, such as the standard hand system. We think this is related to the number of different inputs the system can take on, which just allows for better granularity of the data.

As pediatric orthopaedic surgeons who treat scoliosis, growth is really important to us. Which system should we use? What is your recommendation?
That’s a great question that everyone has asked us. Unfortunately, with our data set, we couldn’t pinpoint exactly one system to use. Given that the discrepancies were pretty minimal, it’s probably okay to use whatever you know [or] existing radiographs we already have in their charts.

Anything else you would like the audience to know about the project?
We are actually working on some exciting advancements in automating skeletal maturity systems, which I think is probably the way of the future. There are a lot of inter- and intra-rater reliability issues with some of these systems, and they can be difficult and tedious to do by hand. Automated reporting of the skeletal maturity systems is our next step and where this is all going.

Selina Poon, MD, MPH, FAAOS, is a pediatric orthopaedic surgeon and director of research at Shriners Children’s Southern California in Pasadena, California.