• Yuan Le, Hal D Kipfer, Shadie S Majidi, Stephanie Holz, and Chen Lin.
• 1 All authors: Department of Radiology and Imaging Science, Indiana University School of Medicine, 950 W Walnut St, R2 E124, Indianapolis, IN 46202.
• AJR Am J Roentgenol. 2014 Sep 1; 203 (3): W307-14.

ObjectiveThe purpose of this article is to evaluate and compare the artifacts caused by metal implants in breast MR images acquired with dual-echo Dixon and two conventional fat-suppression techniques.Subjects And MethodsTwo types of biopsy markers were embedded into a uniform fat-water emulsion. T1-weighted gradient-echo images were acquired on a clinical 3-T MRI scanner with three different fat-suppression techniques-conventional or quick fat saturation, spectrally selective adiabatic inversion recovery (SPAIR), and dual-echo Dixon-and the 3D volumes of artifacts were measured. Among the subjects of a clinical breast MRI study using the same scanner, five patients were found to have one or more metal implants. The artifacts in Dixon and SPAIR fat-suppressed images were evaluated by three radiologists, and the results were compared with those of the phantom study.ResultsIn the phantom study, the artifacts appeared as interleaved bright and dark rings on SPAIR and quick-fat-saturation images, whereas they appeared as dark regions with a thin bright rim on Dixon images. The artifacts imaged with the Dixon technique had the smallest total volume. However, the reviewers found larger artifact diameters on patient images using the Dixon sequence because only the central region was recognized as an artifact on the SPAIR images.ConclusionMetal implants introduce artifacts of different types and sizes, according to the different fat-suppression techniques used. The dual-echo Dixon technique produces a larger central void, allowing the implant to be easily identified, but presents a smaller overall artifact volume by obscuring less area in the image, according to a quantitative phantom study.

27 keywords...

hide…