Discordance among skeletal maturity indicators: sources, interpretation, and clinical implications

Bone age and regional skeletal maturity staging are widely used to identify growth stage, estimate remaining growth, and support orthopedic treatment decisions in children and adolescents. However, skeletal maturity indicators differ in the regions examined, developmental features observed, scoring rules applied, and reference standards used. Consequently, the same child may receive different age estimates or maturity classifications when a different bone group, anatomical region, imaging modality, or assessment method is selected. This narrative review synthesizes evidence from within-image bone-group comparisons, contemporaneous cross-regional assessments, disease and local-intervention studies, mechanistic research, and validation against clinical outcomes. We organize the evidence according to the information represented by each measure and the outcome against which it has been evaluated. Discordance may arise from the imaging and assessment process itself, including modality, scoring method, method version, and population-specific reference standards. It may also reflect regional differences in growth plate responses to systemic signals, prior growth history, and local mechanical conditions. Animal, cellular, and tissue studies support the biological plausibility of site-specific responses, but they do not quantify the magnitude or sequence of regional differences in children. The clinical value of a skeletal maturity indicator must therefore be established separately for the outcome it is intended to inform, such as peak height velocity, adult height, actual growth of a target bone segment, scoliosis progression, or treatment outcome. High correlation between indicators or a small mean difference at the group level does not ensure close agreement within an individual child. Future research should prioritize contemporaneous paired assessments, longitudinal multiregional cohorts, reporting of individual-level differences and stage reclassification, and multicenter external validation. These approaches can help clarify how discordance should be interpreted and whether its magnitude is sufficient to alter a specific clinical judgment.