Geometry of the proximal femur during growth and its contribution to childhood fractures in healthy children

Hip structural analysis (HSA) of proximal femur scans by dual energy X-ray absorptiometry estimates hip geometry and structural strength, but little is known about how these parameters change during peak bone mass development. We describe age-related changes, precision and long-term stability, and sex and race/ethnic identity differences in hip geometry measures by HSA, and test whether HSA measures predict childhood fractures in healthy children. We used data from the Bone Mineral Density in Childhood Study, a multi-center, longitudinal cohort study of 2,014 healthy U.S. children, ages 5-23y, with up to 7 annual visits. HSA measures included cortical thickness, cross-sectional area, cross-sectional moment of inertia, section modulus, buckling ratio, and bone width at the narrow neck and femoral shaft, and hip axis length. Additional measures included self-identified race and ethnicity, Z-scores for height, BMIZ, appendicular lean soft tissue mass index and fat mass index, and self-reported physical activity, calcium intake, Tanner stage and fractures. Results indicated age-related trends and sex differences in hip geometry. Reference ranges were generated and HSA Z-scores were adjusted for height-for-age Z-score. Femoral shaft measures showed better precision (CV: 1.1 to 3.7%) than narrow neck measures (CV: 2.2 to 7.4%); buckling ratio (both sites) was the least precise. HSA Z-scores tracked well over one year (0.77 to 0.94). Narrow neck buckling ratio (HR 1.16 [95% CI: 1.02, 1.31]) and hip axis length Z-scores (HR 1.20 [95% CI: 1.00, 1.44]) associated with fracture risk. When adjusted for covariates (BMD Z-score, BMIZ, Tanner stage, sex), narrow neck cross-sectional area, cross-sectional moment of inertia, section modulus and bone width, and femoral shaft cross-sectional area and section modulus Z-scores positively associated with fractures (e.g., section modulus Z-score associated with 44% increased fracture risk). These findings provide the foundation for evaluating hip geometry as an indicator of bone strength in children.