Dosimetric impact of metal artifact reduction for spinal implants in stereotactic body radiotherapy

Abstract Background Metal artifacts due to spinal implants can affect the accuracy of dose calculation for radiotherapy. However, the dosimetric impact of metal artifact reduction for spinal implants in Stereotactic Body Radiotherapy (SBRT) plans has not been well studied. The purpose of this study was to evaluate the dosimetric impact of a commercial metal artifact reduction method for different commercial dose calculation algorithms in spinal SBRT planning. Methods A gammex phantom and 10 clinical computed tomography (CT) images were selected to investigate the effects of titanium implants. To achieve metal artefact suppression, a commercial orthopedic metal artifact reduction (OMAR) algorithm was employed. Dose calculations for SBRT were conducted on both uncorrected and artifact-suppressed images using three commercial algorithms (Analytical Anisotropic Algorithm (AAA), Acuros XB(AXB) and Monte Carlo (MC)). Dose discrepancies between artifact-uncorrected and artifact-suppressed cases were appraised using Dose Volume histograms (DVH) and 3D gamma analysis with different distance to agreement (DTA) and dose difference criteria. Gamma agreement index (GAI) was denoted as G(∆D, DTA). Results No significant dose differences between the artifact-uncorrected and artifact-corrected datasets were observed for titanium spinal implants. For all parameters of DVH, average differences did not exceed 0.7% in PTV and 2.08% in PRV SC. Gamma index test yielded G(3,3) ≥ 99.8%, G(2,2) > 99.2% and G(1,1) > 93.8% for all patient cases. Moreover, MC algorithm was found to be more sensitive to the presence of titanium inserts compared to AXB and AAA. Conclusions Dosimetric impact of artifacts caused by titanium implants is not significant in spinal SBRT plan, which indicates that dose calculation algorithms might not be very sensitive to CT number variation caused by titanium inserts. In addition, MC algorithm has higher sensitivity to the presence of metal artifacts than that of AXB and AAA algorithms, indicating that MC is preferable for dose calculation in spinal SBRT planning.