Shape-based profiling of cortical bone mineral density in growing children: a proof-of-concept pQCT study.

Markarian, Anna Maria; Taaffe, Dennis R; Galvão, Daniel A; Wilkie, Jodie Cochrane; Peddle-McIntyre, Carolyn J; Schiferl, Daniel J; Newton, Robert U · Bone · 2026

cross_sectional · Level IV

Where this comes from

Abstract

Understanding how bones adapt during growth is essential for promoting lifelong skeletal health. Peripheral quantitative computed tomography (pQCT) is a widely used imaging technique for assessing volumetric bone mineral density (vBMD) in pediatric populations; however, conventional methods may obscure subtle cortical adaptations. The primary aim of this study was to evaluate the feasibility of shape-based cortical vBMD profiling, with an exploratory objective of assessing its sensitivity to inter-individual variation in lean mass index (LMI). Eighteen children (6-12 years) underwent pQCT scanning at the 38% and 66% tibial sites (distal and proximal diaphysis). Cortical vBMD was quantified across three concentric compartments (endocortical, mid-cortical, and pericortical). Differences in vBMD across compartments were assessed using one-way repeated-measures ANOVA. Quadratic models were fit, and participant-specific linear (slope) and quadratic (curvature) coefficients were extracted to characterize cortical vBMD profiles. Exploratory regression analyses examined associations between these coefficients and dual-energy X-ray absorptiometry-derived LMI. Linear and quadratic coefficients were successfully extracted for all participants at both tibial sites, supporting the feasibility of shape-based cortical profiling. Significant differences in vBMD were observed across cortical compartments (p < 0.001). Quadratic models explained 65-72% of the variance in normalized vBMD, and the extracted shape coefficients captured inter-individual and site-specific variation associated with LMI, suggesting sensitivity to biologically meaningful differences in cortical mineral distribution. Shape-based cortical vBMD profiling using pQCT is feasible in children. With further refinement and validation in larger, diverse pediatric cohorts, this approach may offer a sensitive tool for investigating skeletal adaptations during growth.

Medical subject headings