IMPE Abstracts

Osteogenesis imperfecta (OI) is a hereditary skeletal disease characterized by bone fragility. Bone mineral density (BMD), evaluated by dual-energy X-ray absorptiometry (DXA), is used to assess bone brittleness. The height-for-age Z-score (HAZ)-adjusted BMD-for-age Z-score (BMD_HAZ) to reduce the confounding factor of short stature is calculated in children and adolescents with OI. However, even with BMD_HAZ, severity evaluation in children and adolescents with OI is challenging because of the abnormalities in bone quality that BMD analysis cannot accurately assess. The trabecular bone scores (TBS) and bone mineral apparent density (BMAD), which represent the structural integrity of bone and bone-size-associated BMD, respectively, are associated with fracture risk. Recently, age- and sex-specific reference ranges have been reported, enabling the calculation of Z-scores for children. To evaluate the clinical applicability, we analyzed the association between the Z-scores of TBS and BMAD and BMD_HAZ, bone fragility, and genetic variants. Among all OI patients who underwent DXA in our hospital from 2015 to 2022, 49 patients, aged 5 to 20 years at the time of examination, were enrolled. We retrospectively reviewed their clinical features and DXA data. The annual fracture rate (FR) was used as an index of bone fragility. Multiple regression analysis for FR was performed with the associated variables: age, sex, and Z-scores of BMD_HAZ, BMAD, and TBS (R²= 0.314, P=0.0050) and the BMAD Z-score was significantly negatively affected (standard β coefficient [β] = -0.725, P=0.0002), while the TBS Z-score tended to show a negative effect (β = -0.245, P=0.0876). COL1A1/2 and IFITM5 pathogenic variants were detected in 41 and 1 of 49 patients, respectively. Patients with COL1A1/2 nonsense and frameshift variants (n=17) resulting in haploinsufficiency and mild phenotype, while the glycine substitution variants (n=9) cause severe OI. The haploinsufficient group had significantly less FR than that of the glycine substitution group (P=0.027). To evaluate non-severe subset of OI, we analyzed only individuals with haploinsufficient variants and found that their TBS Z-score was negatively correlated with FR (r= -0.50, P=0.042). On the other hand, in glycine substitution group, FR was negatively correlated with the Z-score of BMAD (r= -0.74, P=0.022). These findings suggest that TBS and BMAD are useful in assessing children and adolescents with OI with specific genetic variants.