Prospective evaluation of MRI artefacts following breast conserving surgery and sentinel lymph node dissection with the magnetic technique

• Published in: European journal of surgical oncology 2025-01, Vol.51 (1), p.109368, Article 109368
• Main Authors: Jazrawi, Allan, Pantiora, Eirini, Abdsaleh, Shahin, Ng, Chin Lian, Zouzos, Athanasios, Gagliardi, Tanja, Wärnberg, Fredrik, Eriksson, Staffan, Karakatsanis, Andreas
• Format: Article
• Language: English
• Subjects: Adult; Aged; Breast cancer; Breast conserving surgery (BCS); Breast Neoplasms - diagnostic imaging; Breast Neoplasms - pathology; Breast Neoplasms - surgery; Cancer and Oncology; Cancer och onkologi; Carcinoma, Intraductal, Noninfiltrating - diagnostic imaging; Carcinoma, Intraductal, Noninfiltrating - pathology; Carcinoma, Intraductal, Noninfiltrating - surgery; Contrast Media; Female; Humans; Kirurgi; Magnetic Iron Oxide Nanoparticles; Magnetic resonance imaging (MRI); Magnetic Resonance Imaging - methods; Mammography - methods; Mastectomy, Segmental - methods; Middle Aged; Prospective Studies; Sentinel lymph node (SLN); Sentinel Lymph Node - diagnostic imaging; Sentinel Lymph Node - pathology; Sentinel Lymph Node Biopsy - methods; Superparamagnetic iron oxide nanoparticles (SPIO; Surgery
• ISSN: 0748-7983; 1532-2157; 1532-2157
• DOI: 10.1016/j.ejso.2024.109368

Superparamagnetic Iron Oxide (SPIO) nanoparticles serve as a promising tracer for sentinel lymph node (SLN) detection in breast cancer. Concerns exist regarding artefacts on postoperative Magnetic Resonance Imaging (MRI), especially following breast conservation (BCS). In this prospective observational study, 97 women with DCIS or invasive breast cancer underwent BCS and SLN detection with peritumoral SPIO injection administered up to four weeks before surgery. Postoperatively, patients were followed with MRI and mammograms, and imaging outcomes were evaluated by four, independent breast radiologists, blinded to clinical data, using a predefined, standardized questionnaire. In 97 patients included, there was inter-rater discordance in the prevalence of “any artefact” (range: 24.1–74.4 %; weighted average: 32.4 %) and “SPIO specific artefact” (range: 12.0–49.4 %; weighted average: 20.9 %). The median area of “any artefact” was 9.24 cm2 (iqr 4.72, 15.50) and SPIO specific artefact 9.88 (iqr 5.32, 15.5). Likert scores indicated higher difficulty interpreting MRI if artefacts were present (OR: 2.295, 95 % CI 1.028, 5.123; p = 0.043), but this was reduced if intravenous contrast was administered (OR: 0.177, 95 % CI 0.091, 0.342; p