Site-Specific Material Decomposition Versus Universal Z-Effective: Optimizing Opportunistic Bone Mineral Density...

Dual-energy computed tomography (DECT) represents a high-precision tool for opportunistic bone mineral density (BMD) assessment, yet optimal algorithmic standardization across anatomical sites remains debated. This opportunistic study evaluated 60 postmenopausal female patients who underwent both dual-energy X-ray absorptiometry (DXA) and contrast-enhanced spectral DECT (80/140 kVp) within a 6-month interval. Quantitative trabecular measurements were obtained at the femoral neck and lumbar spine (L1-L4) across five distinct Base Material Pairs (BMPs) and Effective Atomic Number (Z-Effective). Material decomposition revealed significant anatomical site dependency: fat-based pairs demonstrated superior correlation with DXA BMD at the femoral neck (Calcium-Fat: ρ = 0.703; p ≤ 0.001), whereas water-based pairs performed best in the cellular, red-marrow-rich lumbar spine (CaOxMono-Water: ρ = 0.525; p ≤ 0.001). In contrast, Z-Effective functioned as a robust biomarker, maintaining significant positive linear correlations across both the proximal femur (ρ = 0.652; p ≤ 0.001) and lumbar spine (ρ = 0.489; p ≤ 0.001). Multivariable linear regression demonstrated severe multicollinearity between site-matched BMPs and Z-Effective (VIF > 5.0), confirming they capture identical physical mineral depletion. In conclusion, opportunistic DECT screening requires anatomically tailored BMP selection or, alternatively, Z-Effective alone as a unified spectral parameter to achieve reliable, accurate BMD evaluation during routine CT acquisitions.