Abstract: |
OBJECTIVE. The purpose of this study was to assess the utility of MRI in differentiating benign and malignant causes of hepatic lesions deemed too small to characterize on CT in women with breast cancer. MATERIALS AND METHODS. A list of all women with breast cancer diagnosed between January 2000 and June 2003 was cross-referenced with radiologic and medical records to identify patients who had undergone MRI of the liver and who had had a hepatic lesion too small to characterize reported on previous CT performed after the diagnosis of breast cancer. The cause reported at MRI for each hepatic lesion too small to characterize seen on CT was recorded as benign, malignant, indeterminate, or no lesion seen on MRI. Subsequent relevant imaging examinations and medical records were reviewed. RESULTS. Seventy-six (1.4%) of 5,440 women underwent MRI that included the liver. In 38 (50%) of the women a hepatic lesion too small to characterize was found on CT performed before MRI. The reported MRI diagnoses of hepatic lesions too small to characterize on CT included benign lesions (n = 22 women), lesions that remained indeterminate (n = 11 women), no lesion seen on MRI corresponding to the site of a lesion too small to characterize on CT (n = 8 women), and metastatic lesions (n = 2 women). Four women had more than one reported type of hepatic lesion. Subsequent findings of imaging, biopsy, or both performed on eight of 11 women with indeterminate lesions supported benign diagnoses in all eight women. CONCLUSION. In women with newly diagnosed breast cancer and no definite liver metastasis on initial CT, immediate further evaluation of hepatic lesions too small to characterize with MRI offers only marginal benefit. In only approximately 5% of such women will lesions too small to characterize be shown on MRI to represent metastasis. © American Roentgen Ray Society. |