Evaluation of Skeletal Maturity Using the Distal Femoral Physeal Central Peak is Not Significantly Affected by Radiographic Projection.

BACKGROUND: Accurate estimation of skeletal maturity is important in several pediatric orthopaedic conditions. The current gold standard for estimating skeletal maturity using the Greulich and Pyle Bone Atlas is complex and shown to have significant interobserver variability. Recent data have shown peak height velocity to occur on average at 90% of final adult height, providing an improved gold standard to quantify skeletal maturity, facilitating the investigation of different skeletal maturity systems. Measurement of topographical changes to the developing distal femoral physis on anteroposterior (AP) radiographs allow for calculation of the central peak value (CPV), a quantitative method shown to provide accurate prediction of 90% of final adult height. The purpose of this study was to assess the clinical tolerance of the CPV method to varying beam angles by comparing measurement reliability between AP radiographs of the knee versus standing hip-to-ankle leg-length radiographs.

METHODS: We searched our institution's pediatric orthopaedic clinical database for skeletally immature patients evaluated with both standard AP radiographs of the knee as well as standing hip-to-ankle radiographs. Patients included female individuals aged 7 to 16 years and male individuals aged 7 to 18 years with both radiographs within 6 months. CPV was measured using a previously published method. Intraclass correlation coefficient was calculated to determine the level of agreement between observers in all available radiographs. CPVs between AP radiographs of the knee and standing hip-to-ankle radiographs were compared using a paired t test to determine if there is a significant difference between radiographic projection and sex.

RESULTS: A total of 78 subjects meeting appropriate inclusion and exclusion criteria were identified. intraclass correlation coefficient value was 0.873, indicating excellent interobserver reliability for CPV measurements. The mean time between radiographs was 0.30 years for male and 0.27 years for female patients. CPV values between the 2 radiographic projections were not significantly different in male (P=0.37), female (P=0.22) or male+female patients (P=0.17). CPV values were significantly higher in male patients on both AP radiographs (P<0.001) and standing hip-to-ankle radiographs (P<0.001) when compared with female patients.

CONCLUSIONS: The CPV is a quick, quantitative method for estimating skeletal maturity. CPVs are not significantly different between standard AP radiographs of the knee versus standing hip-to-ankle leg-length radiographs, expanding the potential to utilize this method without the need for additional expense or radiation.

LEVEL OF EVIDENCE: Level III-retrospective comparative study.