Updated breast MRI lexicon

Abstract

The American College of Radiology (ACR) BI-RADS Breast Imaging Lexicon was developed over a 5 year period, initially organized in 1999 by a Breast MRI “Lesion Diagnosis Working Group” and later by the ACR. This work resulted in a lexicon (dictionary) of terms to describe breast lesions on contrast-enhanced breast magnetic resonance imaging studies. The first publication was developed when breast MRI was first being used and referencing the little data available to produce the lexicon of terms in describing lesions seen on breast MRI. Even though the ACR Lexicon has proved robust for the most part, since its publication there have been technical advances in breast MRI hardware, coils and pulse sequences, resulting in new information on breast cancer morphology and kinetic behavior. Publication on MRI studies using the new hardware and software resulted in increases in cancer sensitivity and specificity in published literature. As a result, information is now available on technical parameters required to obtain both high spatial and temporal resolution on breast MRI scans that resulted in high sensitivity and specificity. These advances and scientific data were used in developing and updating the new Lexicon. The new breast MRI Lexicon is chaired by Elizabeth A Morris, M.D and is expected to be published in the fall of 2012. The new ACR Lexicon includes a web-based format (hardcopy is also available), recommends mammography/US/MRI correlation, and has evidence based information with hyperlinks to references. The Lexicon also includes sections on Quality Assessment with technical parameters and kinetic/functional considerations that influence MRI images. There is a new section on audit recommendations. The ACR also lists a separate ACR Breast MRI Accreditation Program that accredits breast MRI facilities in the United States with technical, outcome and personnel credentialing criteria. Breast MRI Lexicon recommendations include adding T2weighted non-contrast sequences, and combined reporting of MRI with mammographic and ultrasound findings. Finally, simultaneous bilateral MRI scans are recommended and are considered standard of care. There is a new section on breast implants, with a lexicon describing the implant type, location, and findings commonly seen in intact and ruptured implants. In silicone breast implant studies, imaging terms include normal radial folds, subcapsular line, linguine and keyhole/teardrop. Importantly, there is a new section describing background parenchymal enhancement (BPE) that describes how much of the normal breast parenchyma enhances, which in turn can influence the sensitivity of breast MRI to detect cancer. The BPE can be described as none, minimal, mild, moderate or marked. Terms that previously were used to describe findings seen on breast MRI such as “diffuse stippled enhancement” may actually represent moderate BPE, which is a normal finding. As before, both morphology and kinetic information is considered important. There are no changes to the kinetic terminology. Morphology terms in the lexicon have been added or deleted, adding terms that were commonly used, and deleting those that were not used in the scientific literature. Deleted terms include central and septal enhancement, and enhancing septations. Morphology additions include clustered ring enhancement in contrast-enhanced studies. Morphology terms that have been revised include “non-masslike” to “non-mass” enhancement, and “irregular” mass margins to “uneven” mass margins in masses that have an irregular shape. Last, the Lexicon recommends a final BIRADS assessment including BIRADS codes 0 – Incomplete, Need Additional Imaging Evaluation, 1 – Negative, 2 – Benign, 3 – Probably Benign, 4 – Suspicious, 5 – Highly Suggestive of Malignancy, and 6 – Known Biopsy Proven Malignancy. The ACR Breast MRI Lexicon publication also includes an atlas that illustrates the key findings described in the lexicon. Because the contributors to the atlas were based both in the USA and around the world, there are a wide variety of techniques and pulse sequences that were used to generate the images. Variability in the examples include axial, sagittal, multi-planar reconstructions, fatsuppressed and non-fat-suppressed images chosen as state-of-theart, providing the reader with the rare chance to view MRI studies of specific findings using high-quality, varying techniques.