OBJECTIVE: This study was undertaken to introduce a new breast magnetic resonance (MR) imaging technique, 3-phase dynamic MR imaging with 2-way subtraction, and to examine the morphologic and kinetic features of malignant and benign breast lesions using this technique. METHODS: In 99 breasts from 89 consecutive women with suspicious breast lesions (age range: 32-72 years, mean = 48.2 years), MR imaging was performed using a fat-saturated, sagittal, gradient-echo sequence in 3 phases (1 precontrast and 2 postcontrast scans). Two-way subtraction was performed: standard subtraction (early postcontrast minus precontrast scans) and reverse subtraction (early postcontrast minus late postcontrast scans). A radiologist interpreted all the images, described the breast lesions (using the Breast Imaging Reporting and Data System MR lexicon), and assessed the lesions prospectively as being benign or malignant. The lesions with at least 1 malignant feature were classified as malignant. The MR classifications were compared with the biopsy or follow-up results. RESULTS: Sixty-five (65.7%) breast lesions were malignant, and 34 (34.3%) were benign. Forty-two (97.7%) of the 43 cases of microinvasive or invasive ductal cancer showed malignant morphologic features, and 38 (88.4%) showed washout kinetics. Eighteen (81.8%) of the 22 cases of ductal carcinoma in situ showed malignant morphologic features, and 11 (50.0%) showed washout kinetics. Twenty-nine (85.3%) of the 34 benign lesions showed neither malignant morphologic features nor washout kinetics. The sensitivity and specificity were 92.3% and 91.2% by morphologic analysis and 93.8% and 85.3% by the combined method (morphology plus kinetic analysis), respectively. CONCLUSIONS: This study demonstrated that 3-phase dynamic breast MR imaging with 2-way subtraction and a maximum intensity projection (MIP) image is a simple and useful technique for identifying breast lesions. Although the addition of the kinetic criteria enhanced the sensitivity at the cost of specificity, radiologists can diagnose a malignant lesion showing non-mass-like enhancement more confidently using reverse subtraction imaging. Therefore, this study proposes 3-phase dynamic imaging with 2-way subtraction and an MIP image as one of the standard protocols of breast MR imaging.
OBJECTIVE: This study was undertaken to introduce a new breast magnetic resonance (MR) imaging technique, 3-phase dynamic MR imaging with 2-way subtraction, and to examine the morphologic and kinetic features of malignant and benign breast lesions using this technique. METHODS: In 99 breasts from 89 consecutive women with suspicious breast lesions (age range: 32-72 years, mean = 48.2 years), MR imaging was performed using a fat-saturated, sagittal, gradient-echo sequence in 3 phases (1 precontrast and 2 postcontrast scans). Two-way subtraction was performed: standard subtraction (early postcontrast minus precontrast scans) and reverse subtraction (early postcontrast minus late postcontrast scans). A radiologist interpreted all the images, described the breast lesions (using the Breast Imaging Reporting and Data System MR lexicon), and assessed the lesions prospectively as being benign or malignant. The lesions with at least 1 malignant feature were classified as malignant. The MR classifications were compared with the biopsy or follow-up results. RESULTS: Sixty-five (65.7%) breast lesions were malignant, and 34 (34.3%) were benign. Forty-two (97.7%) of the 43 cases of microinvasive or invasive ductal cancer showed malignant morphologic features, and 38 (88.4%) showed washout kinetics. Eighteen (81.8%) of the 22 cases of ductal carcinoma in situ showed malignant morphologic features, and 11 (50.0%) showed washout kinetics. Twenty-nine (85.3%) of the 34 benign lesions showed neither malignant morphologic features nor washout kinetics. The sensitivity and specificity were 92.3% and 91.2% by morphologic analysis and 93.8% and 85.3% by the combined method (morphology plus kinetic analysis), respectively. CONCLUSIONS: This study demonstrated that 3-phase dynamic breast MR imaging with 2-way subtraction and a maximum intensity projection (MIP) image is a simple and useful technique for identifying breast lesions. Although the addition of the kinetic criteria enhanced the sensitivity at the cost of specificity, radiologists can diagnose a malignant lesion showing non-mass-like enhancement more confidently using reverse subtraction imaging. Therefore, this study proposes 3-phase dynamic imaging with 2-way subtraction and an MIP image as one of the standard protocols of breast MR imaging.