Assessing Skeletal Maturity: Searching for the Holy Grail: Commentary on an article by Ryan J. Furdock, MD, et al.: "Systematic Isolation of Key Parameters for Estimating Skeletal Maturity on Anteroposterior Wrist Radiographs".

Abstract

Commentary Ryan J. Furdock and his fellow researchers have produced a highly sophisticated study focused on a topic of wide interest to the orthopaedic community: simplified and improved assessment of skeletal maturity. A number of existing methods of estimating skeletal maturity were assessed in this study: the Fels hand-wrist method, the Greulich and Pyle atlas, the Sanders hand system, and the distal radius and ulna classification. Using a subset of data from the Bolton-Brush growth study (80 pediatric patients with serial radiographs up to skeletal maturity), they produced what they have termed a “modified Fels wrist skeletal maturity system” (named for businessman and philanthropist Samuel Simeon Fels [1860-1950]) that takes a dizzying list of radiographic parameters and boils them down to the key 8 (7 from the Fels system and 1 from the Greulich and Pyle system). The psychometrics of the modified system are quite favorable, with interrater and intrarater reliability of 0.79 to 0.97 in addition to the best precision and the least outlier estimates as compared with the other methods. The authors also mention a free mobile phone application considered to streamline use of the modified Fels wrist system. The authors acknowledge inherent limitations of the Bolton-Brush data set, which was collected from a predominantly Caucasian population in northern Ohio from 1929 to 1942, and suggest that race and sex-based corrections may be necessary to maximize accuracy. The concern about essentially mono-racial data from nearly 100 years ago is not new and criticism has usually focused on the generalizability of these data to current multi-ethnic/multi-racial populations. However, I would submit that another concern is whether skeletal maturation has measurably changed in 100 years. Fels longitudinal growth data (which ironically also dates back to 1920s Ohio—starting in Yellow Springs, Ohio, in 1929) has recently been further assessed by researchers at the University of Missouri, who compared children born in the 1930s to those born in the 1990s1. Bone maturation was noted almost 1 full year earlier in the 1990s group. These changes in bone maturation have also been shown to vary according to anatomic area and sex as well as tempo and duration2,3. In pediatric orthopaedics, an endless array of syndromes and conditions (e.g., myelomeningocele) with their own skeletal growth quirks may also cause us to question Bolton-Brush-derived data4. Clearly, more work remains to be done. The simplicity and accuracy of what Furdock et al. are offering are certainly attractive, but as the authors themselves have stated, “the system needs external validation in a modern, diverse pediatric cohort.”