Dosimetric impact of titanium cranioplasty in Gamma Knife radiosurgery: a technical note comparing TMR 10 and convolution algorithms

Abstract

To evaluate the dosimetric impact of a titanium cranioplasty implant in Gamma Knife (GK) radiosurgery by comparing the TMR 10 algorithm with the CT density-corrected Convolution algorithm. Methods Paired pre- and post-cranioplasty CT datasets from a single patient were analyzed. A 3.5 cc virtual target was placed at five locations relative to the titanium hardware. Baseline plans were optimized on the pre-cranioplasty CT using the Convolution algorithm (20 Gy to the 50% isodose line), then transferred to the post-cranioplasty CT via rigid registration and recalculated with both Convolution and TMR 10 without re-optimization. Endpoints included minimum, mean, and maximum dose; Paddick Conformity Index (PCI); and DVH percentiles (D98%, D50%, D2%). Results Plan quality was preserved across all locations and algorithms. Maximum dose (33.3–34.5 Gy) and PCI (0.79–0.82) remained stable. Post-cranioplasty recalculation yielded small increases in minimum point dose (mean + 0.10 Gy) and mean dose (mean + 1.06 Gy). Differences between TMR 10 and Convolution were modest and most pronounced beneath the implant (up to + 0.30 Gy in D50%). Conclusion Titanium cranioplasty caused small, spatially limited dosimetric perturbations in GK radiosurgery, with conformity and coverage metrics remaining largely stable. CT density-corrected Convolution is preferable when targets or critical structures are adjacent to metallic hardware, and dose metrics near the metal-tissue interface warrant careful scrutiny.