PURPOSE: To determine whether the presence of extensive intraductal components (EIC) in breast carcinomas can be accurately evaluated on magnetic resonance (MR) images. MATERIAL AND METHODS: Ninety-three women with breast cancer, aged between 32 and 79 years (mean 54 years), underwent three-dimensional dynamic MR imaging (dyMRI) with fat suppression and magnetization transfer contrast before breast-conserving surgery. The tumors were classified on dyMRI as circumscribed, microlobulated, and/or speculated, and their size was measured. Spotty or linear continuous enhancement (SLE) from the main tumor to the nipple and segmental enhancement surrounding the main tumor (SE) were considered indicative of intraductal tumor spread. The correlation between preoperative MRI and macroscopic and microscopic findings was examined. RESULTS: On MR images, the tumor sizes ranged from 0.8 to 3.4 cm. These measurements coincided with histologic measurements in circumscribed tumors. However, in tumors with microlobulated or spiculated borders, tumor size tended to be underestimated on MR images. Of 93 patients, 59 (63.4%) had histologically confirmed EIC; 42 of the 59 cancers (71.2%) manifested SLE or SE on MR images. The sensitivity, specificity, and accuracy of MR imaging in detecting EIC were 71%, 85%, and 76%, respectively. CONCLUSION: MR imaging facilitates the detection of EIC in breast masses. This information is valuable for the planning of breast-conserving surgery.
PURPOSE: To determine whether the presence of extensive intraductal components (EIC) in breast carcinomas can be accurately evaluated on magnetic resonance (MR) images. MATERIAL AND METHODS: Ninety-three women with breast cancer, aged between 32 and 79 years (mean 54 years), underwent three-dimensional dynamic MR imaging (dyMRI) with fat suppression and magnetization transfer contrast before breast-conserving surgery. The tumors were classified on dyMRI as circumscribed, microlobulated, and/or speculated, and their size was measured. Spotty or linear continuous enhancement (SLE) from the main tumor to the nipple and segmental enhancement surrounding the main tumor (SE) were considered indicative of intraductal tumor spread. The correlation between preoperative MRI and macroscopic and microscopic findings was examined. RESULTS: On MR images, the tumor sizes ranged from 0.8 to 3.4 cm. These measurements coincided with histologic measurements in circumscribed tumors. However, in tumors with microlobulated or spiculated borders, tumor size tended to be underestimated on MR images. Of 93 patients, 59 (63.4%) had histologically confirmed EIC; 42 of the 59 cancers (71.2%) manifested SLE or SE on MR images. The sensitivity, specificity, and accuracy of MR imaging in detecting EIC were 71%, 85%, and 76%, respectively. CONCLUSION: MR imaging facilitates the detection of EIC in breast masses. This information is valuable for the planning of breast-conserving surgery.