Assessment of vertebral bone quality in women with type 2 diabetes mellitus using digital tomosynthesis

Type 2 diabetes mellitus (T2DM) increases vertebral fracture risk. Despite normal (or higher) bone mineral density (BMD), patients with T2DM typically have a lower trabecular bone score (TBS), indicating degraded bone microstructure in T2DM. However, assessment of vertebral microstructure using TBS remains limited because it is based on 2D DXA imaging. Due to high in-plane resolution and 3D imaging capabilities, vertebral microstructure and geometry can be quantified using digital tomosynthesis (DTS) imaging. This study aimed to examine the differences in DTS-derived vertebral parameters between T2DM patients and controls without T2DM (NT2DM).

Fifty postmenopausal women with T2DM and 54 controls were recruited. Lumbar spine BMD and trabecular bone score (TBS) were calculated from DXA images. Microstructural (fractal dimension, lacunarity, scale-dependent lacunarity, anisotropy, and line fraction deviation) and geometrical (vertebral area and width) parameters were calculated from DTS images for T12 and L1 vertebrae. Group differences were examined within the framework of generalized linear models.

T2DM and NT2DM were different in race (p < 0.001) but not in age or BMI. T2DM had higher BMD than NT2DM (p < 0.02), but the difference was not present after adjustment for demographic variables (age, BMI, and race) (p > 0.1). T2DM had lower vertebral area (p < 0.002), higher anisotropy (p < 0.03), and marginally lower lacunarity (p = 0.063) as assessed by DTS, even after adjusting for demographic variables.

DTS imaging of vertebrae reveals distinct bone size and microstructure characteristics in patients with T2DM, and thus may be clinically useful to explain increased fracture risk in T2DM beyond DXA.