Room: AAPM ePoster Library
Purpose: To determine the dosimetric impact of a cardiac implantable electronic device (CIED) deactivation magnet during total body irradiation (TBI).
Methods: A Medtronic deactivation magnet with a custom brass insert to shield the central part of the CIED was investigated. To characterize general attenuation capability, 200cGy (6MV, 10×10cm²) was delivered using a Varian linear accelerator to a depth of maximum dose (dmax) in solid water with the magnet and insert centered on the surface. Gafchromic EBT3 film measured surface and dmax dose. Additionally, a TBI treatment was delivered to a RANDO phantom using our institution’s VMAT technique. The technique uses anteroposterior-posteroanterior 6MV arcs, an extended source-to-surface distance, and custom couch with beam spoiler. The phantom received 200cGy with the magnet and insert in place for the entire delivery. Depth dose in three axial slices and surface dose under the magnet were measured with film.
Results: In solid water, the mean dose at dmax was reduced by 19% (161±3cGy) under the magnet and 31% (137±3cGy) under the insert compared to delivery without the magnet and insert (198±4cGy). Mean surface dose was increased to 156±4cGy and 128±3cGy under the magnet and insert respectively, compared to 44±1cGy without the magnet and insert. RANDO dose under the magnet was heterogeneously distributed due to arc delivery, with low and high dose regions located superiorly and inferiorly, respectively. Dose to the body and CIED under the magnet and insert ranges between 148cGy and 203cGy, which spans 74-102% of the prescribed fraction dose.
Conclusion: Using a deactivation magnet and brass shielding insert during TBI has a notable dosimetric impact based on the expected TBI dose homogeneity of ±10%. Additional shielding may be required to meet patient-specific CIED dose constraints. Clinical implications of the non-uniform dose distribution will be dependent on positioning of the CIED and beam geometry.
Funding Support, Disclosures, and Conflict of Interest: This work was supported by the Natural Sciences and Engineering Research Council of Canada and the University of Calgary Eyes High Strategy. There are no relevant financial disclosures or conflicts of interest to declare.