Can MR imaging contribute in characterizing well-circumscribed breast carcinomas?

Research output: Contribution to journalArticle

32 Citations (Scopus)

Abstract

Most well-circumscribed breast masses are benign lesions such as cysts, fibroadenomas, and intramammary lymph nodes. Nevertheless, 10%-20% of breast malignancies are well-circumscribed masses, and these malignancies include papillary, mucinous, medullary, and metaplastic carcinomas, as well as malignant phyllodes tumors. Therefore, it is important to differentiate these well-circumscribed breast malignancies from benign breast lesions, but it is not easy to do so with conventional imaging modalities such as mammography and ultrasonography (US). As an emerging adjunctive imaging method, magnetic resonance (MR) imaging has substantial potential in characterizing well-circumscribed breast carcinomas. Analysis of the lesion signal intensity on nonenhanced T2-weighted MR images, determination of the enhancement pattern, and kinetic curve assessment can greatly help differentiate malignant from benign well-circumscribed breast lesions. Therefore, breast MR imaging can play a substantial role in distinguishing between well-circumscribed benign and malignant breast lesions, especially in cases that are difficult to diagnose by using conventional imaging. In this article, the MR imaging findings of the subtypes of well-circumscribed malignant breast lesions-intracystic papillary carcinoma, invasive papillary carcinoma, mucinous carcinoma, medullary carcinoma, metaplastic carcinoma, and malignant phyllodes tumor-are described and correlated with the histopathologic, mammographic, and US findings.

Original languageEnglish
Pages (from-to)1689-1703
Number of pages15
JournalRadiographics
Volume30
Issue number6
DOIs
Publication statusPublished - 2010 Oct

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging

Fingerprint Dive into the research topics of 'Can MR imaging contribute in characterizing well-circumscribed breast carcinomas?'. Together they form a unique fingerprint.

  • Cite this