Background: The widespread use of mammographic screening has led to increased detection of small tumors that are often difficult to diagnose with conventional imaging modalities such as mammography and ultrasonography. Intraductal spread of breast cancer, a principle risk factor for local recurrence, is also difficult to diagnose with mammography and ultrasonography. We investigated the clinical usefulness of magnetic resonance imaging of the breast in the therapy of breast cancer and we compared it with mammography and ultrasonography.
Study design: A total of 183 patients with primary breast cancer underwent surgery at our institute between September 1, 1999, and November 30, 2002. They were examined preoperatively with magnetic resonance imaging, mammography, and ultrasonography. Magnetic resonance imaging evaluation included contrast-enhanced dynamic studies using IV injection of gadolinium-diethylenetriamine pentaacetic acid.
Results: Detection rates of breast cancers by magnetic resonance imaging, mammography, and ultrasonography were 93.7%, 84.6%, and 97.3%, respectively (magnetic resonance imaging versus mammography, p < 0.05). Patterns of time-intensity curves in dynamic magnetic resonance imaging differed with histologic types. Sensitivity, specificity, and accuracy of detection of intraductal spread were 66.7%, 64.2%, and 65.6% with MRI; 22.2%, 85.7%, and 50% with mammography; and 20.6%, 85.2%, and 50% with ultrasonography, respectively (sensitivity, specificity, and accuracy; p < 0.05, respectively).
Conclusions: Magnetic resonance imaging can diagnose breast cancer as accurately as ultrasonography and more accurately than mammography. Patterns of time-intensity curves correlated with tumor histology. In addition, magnetic resonance imaging can detect intraductal spread more accurately than the other two methods. Magnetic resonance imaging appears to be indispensable in breast-conserving surgery to minimize local recurrence.