The utility of a novel proximal femur maturity index for staging skeletal growth in patients with idiopathic scoliosis

• Main Authors: Cheung, Prudence Wing Hang, Canavese, Federico, Chan, Chris Yin Wei, Wong, Janus Siu Him, Shigematsu, Hideki, Luk, Keith Dip Kei, Cheung, Jason Pui Yin
• Format: Article
• Published: Lippincott Williams & Wilkins 2022
• Subjects: RZ Other systems of medicine; Osteopath
• Online Access: http://eprints.um.edu.my/42897/
• Institution: Universiti Malaya

Background: For growing patients, it is ideal to have a growth plate visible in routine radiographs for skeletal maturity assessment without additional radiation. The proximal femoral epiphyseal ossification is in proximity to the spine; however, whether it can be used for assessing a patient's growth status remains unknown. Methods: Two hundred and twenty sets of radiographs of the spine and the left hand and wrist of patients with idiopathic scoliosis were assessed for skeletal maturity and reliability testing. Risser staging, Sanders staging (SS), distal radius and ulna (DRU) classification, the proximal humeral ossification system (PHOS), and the novel proximal femur maturity index (PFMI) were used. The PFMI was newly developed on the basis of the radiographic appearances of the femoral head, greater trochanter, and triradiate cartilage. It consists of 7 grades (0 to 6) associated with increasing skeletal maturity. The PFMI was evaluated through its relationship with pubertal growth (i.e., the rate of changes of standing and sitting body height BH] and arm span AS]) and with established skeletal maturity indices. Longitudinal growth data and 780 corresponding spine radiographs were assessed to detect peak growth using receiver operating characteristic (ROC) curve analysis. Results: The PFMI was found to be correlated with chronological age (tau(b) = 0.522), growth rates based on standing BH (tau(b) = -0.303), and AS (tau(b) = -0.266) (p < 0.001 for all). The largest growth rate occurred at PFMI grade 3, with mean standing BH growth rates (and standard deviations) of 0.79 +/- 0.44 cm/month for girls and 1.06 +/- 0.67 cm/mo for boys. Growth rates of 0.12 +/- 0.23 cm/mo (girls) and 0 +/- 0 cm/mo (boys) occurred at PFMI grade 6, indicating growth cessation. Strong correlations were found between PFMI gradings and Risser staging (tau(b) = 0.743 and 0.774 for girls and boys), Sanders staging (tau(b) = 0.722 and 0.736, respectively), and radius (tau(b) = 0.792 and 0.820) and ulnar gradings (tau(b) = 0.777 and 0.821), and moderate correlations were found with PHOS stages (tau(b) = 0.613 and 0.675) (p < 0.001 for all). PFMI gradings corresponded to as young as SS1, R4, U1, and PHOS stage 1. Fair to excellent interrater and intrarater reliabilities were observed. PFMI grade 3 was most prevalent and predictive for peak growth based on ROC results. Conclusions: The PFMI demonstrated clear pubertal growth phases with satisfactory reliability. Grade 3 indicates peak growth and grade 6 indicates growth cessation.