Breast MR imaging

Breast cancer is a significant health care problem in the United States. More than 180,000 American women are diagnosed with breast cancer each year, and approximately 50,000 of these women will die of their disease [1]. Breast cancer is the second leading cause of cancer death among women. Imaging plays a crucial role in all aspects of breast cancer care. This includes early detection through screening, diagnosis and associated image-guided biopsy, treatment planning, and follow-up. The limitations of current x-ray mammography have led to extensive efforts over the past 15 years to develop complimentary imaging techniques to improve breast imaging performance, particularly in the radiographically dense breast. The most accepted adjunct modality is breast sonography, which is now widely used in the diagnostic evaluation of women with abnormal screening mammography or clinical exams. Other techniques that have been proposed that are less widely used include scintography with Tc99 Sestamibi and (18)[F]-fluorodeoxyglucose positron emission tomography. Early in its history, MR imaging was proposed as a technique to assist in the detection and diagnosis of breast cancer. Early reports of breast MR imaging confirmed that high-quality MR images of the breast could be obtained with local surface coils [2,3]. In the absence of exogenously injected contrast agents, however, it was difficult to detect breast cancer. In 1989, Kaiser and Zeitler [4] and Heywang et al [5] independently published on the application of MR imaging contrast agents to detect breast cancer. Their findings-that the use of intravenous gadolinium chelates allow MR imaging to detect and possibly diagnose breast cancer-were extremely exciting and led to significant follow-up work. Using a higher resolution, three-dimensional (3D) technique, Harms et al [6] demonstrated the power of MR imaging to detect mammographically and clinically occult breast cancer. They performed careful correlations between examinations performed on women prior to mastectomy and the resultant pathology, and showed that high-resolution MR imaging, performed with a technique pioneered in his laboratory rotating delivery of excitation off resonance (RODEO), was able to detect occult multifocal cancer in up to 40% of women. Despite the extremely high sensitivity for breast cancer MR imaging demonstrated in these early studies, it was clear that contrast enhancement alone was not specific for breast cancer. This led to intense efforts to identify distinguishing characteristics between benign enhancing lesions and malignant enhancing lesions. The use of information obtained from the architecture of the enhancing lesion and qualitative and quantitative interpretations of the pharmicokinetics of enhancement have been studied extensively for this purpose. Although many different pharmacokinetic imaging approaches have been described, the most commonly adopted method is the qualitative approach to the time signal intensity curve of gadolinium, which was popularized by Kuhl et al [7] and Kaiser et al [4]. Work by Orel et al [8] and Nunes et al [9] clearly demonstrated the importance of lesion architecture for distinguishing between benign and malignant lesions. Today most practitioners agree that a combination of gadolinium pharmokinetics and lesion architecture is important for proper interpretation of breast MR images. Recently, breast MR imaging has become more widely used as a supplemental imaging modality in the breast. The most common clinical indications for which breast MR imaging is being performed include difficult diagnostic evaluations and the evaluation of extent of disease within a breast affected with breast cancer. This article discusses breast MR imaging techniques, interpretation strategies, and strategies for clinical implementation.