MR-compatible and conventional localization wires in breast diagnosis: experimental studies on the dislocation and artefact size in MRI.

Purpose: Localization of non-palpable suspicious breast lesions with tumor localization wires is commonly used before surgical excision. Dislocation o f the wire is possible. The possibility of dislocation for different types of tumor localization wires was examined experimentally. Methods: 22 differ ent tumor localization wires were inserted into the subcutaneous fat-tissue of a pig. The wire tip was recorded under fluoroscopy. Traction was applie d to the wire using a spring-balance. The artifact size of MR-compatible wi res was examined in a 1.0 Tesla MR imager. Results: The localization wires with an X-configuration and long hook could not be dislocated by a force up to the maximum of 12 Newton. The retractable localization wires were easil y dislocated at traction forces of 1 Newton. Other wires with only one hook could withstand a traction from 2 up to 7 Newton. The wires with a Z-confi guration and wires with short hooks showed a highly variable rate of resist ance. The artifact size of localization systems was between 10.1 and 11.0 m m for gradient-echo sequences and between 6.7 and 7.1 mm for spin-echo sequ ences. Conclusion: Tumor localization wires with an X-configuration and lon g hooks were best anchored in the tissue. The advantage of repositioning re tractable wires is confined by the high risk of accidental dislocation. No reliable fixation in the tissue could be achieved using wires with a Z-conf iguration and wires with short hooks. MR-compatible wires were comparable t o non-MR-compatible wires concerning the rate of resistance. The size of th e artifacts of MR-compatible systems showed no relevant differences.