Background and Purpose: This study investigated depth-dose alterations caused by bone cement in solid phantoms during vertebroplasty. Dose variations of megavoltage (MV) photon beams after traversing high-density bone cement were measured to estimate the effective density of bone cement. Methods: A phantom mimicking human anatomy was fabricated with 2.5-cm acrylic embedded with bone cement. An Elekta VersaHD linear accelerator was used with 6 MV and 6 MV flattening filter-free (FFF) photon beams, field sizes ranging from 3 × 3 cm 2 to 10 × 10 cm 2 . Depth dose, planar dose, and point dose were measured using an Advanced Markus chamber, PTW 1600 two-dimensional array, and Semiflex three-dimensional ion chamber. These were compared with calculated doses from treatment plans with various overridden density values for bone cement. Results: Minimal differences in percent depth dose were observed after MV photons (6 MV and 6 MV FFF) penetrated the bone cement. At 1 cm beyond cement, measured versus calculated doses differed by 1.2% to 2.7%. Within 1 to 5 cm behind cement, dose discrepancies were within 3% to 5%. At 3-cm depth behind cement, point-dose differences measured by the Semiflex chamber indicated smaller deviations at lower override densities, with an optimal match at 0.6 g/cm 3 . The two-dimensional array showed near-perfect gamma passing rates (∼100%) at an overridden density of 0.8 g/cm 3 , demonstrating the smallest center-point dose discrepancies. A higher override density of 1.4 g/cm 3 yielded results similar to the plan without density override, suggesting minimal advantage at higher override densities. Conclusion: This study revealed that dose perturbations induced by bone cement caused the treatment planning system to underestimate actual delivered doses by approximately 3% to 5%. Optimal density override values for bone cement in treatment planning systems appear to be around 0.6 to 0.8 g/cm 3 . Clinical trials and further data acquisition are necessary to validate precise density values due to variability in cement composition and incomplete cement infusion.
Wu et al. (Tue,) studied this question.