INTRODUCTION: For patients with Paget disease (PD) of the nipple, preoperative imaging to detect and evaluate the extent of an underlying malignancy can facilitate appropriate treatment planning. The purpose of this study was to evaluate the role of breast MRI in this setting. STUDY DESIGN: Using ICD-9 codes for "breast symptoms," we identified 2,294 patients without a current cancer diagnosis seen at our institution (1995 to 2006). Sixty-nine patients (3%) had nipple changes suspicious for PD as the only physical finding. Skin/nipple biopsy confirmed PD in 39 of 69 (57%) patients. Thirty-four patients were eligible for review. RESULTS: Surgical pathology identified cancer in 32 of 34 (94%) patients (7 invasive ductal carcinoma, 25 ductal carcinoma in situ). Nineteen (59%) cancers were confined to the central quadrant of the breast (unifocal). Preoperative imaging (mammography 34 of 34, MRI 13 of 34) detected 15 of 32 (49%) cancers. Mammography detected 11 cancers, accurately demonstrating extent of disease in 9 of 11 patients. MRI detected seven cancers, accurately demonstrating extent of disease in six of seven patients. After positive mammography (n = 11), MRI (n = 5) did not change management. After negative mammography (n = 23), MRI (n = 8) detected otherwise occult disease in 4 of 8 patients, accurately demonstrating extent of disease in 4 of 4 patients and ruling out an underlying cancer in 1 of 1 patient. CONCLUSIONS: Ninety-four percent of patients with biopsy-proved PD as the only physical finding had an underlying cancer and 59% had unifocal disease. Negative preoperative imaging did not reliably exclude an underlying cancer, but the increased sensitivity of MRI detected otherwise occult disease. In the setting of negative mammography, MRI can facilitate treatment planning for patients with PD.
INTRODUCTION: For patients with Paget disease (PD) of the nipple, preoperative imaging to detect and evaluate the extent of an underlying malignancy can facilitate appropriate treatment planning. The purpose of this study was to evaluate the role of breast MRI in this setting. STUDY DESIGN: Using ICD-9 codes for "breast symptoms," we identified 2,294 patients without a current cancer diagnosis seen at our institution (1995 to 2006). Sixty-nine patients (3%) had nipple changes suspicious for PD as the only physical finding. Skin/nipple biopsy confirmed PD in 39 of 69 (57%) patients. Thirty-four patients were eligible for review. RESULTS: Surgical pathology identified cancer in 32 of 34 (94%) patients (7 invasive ductal carcinoma, 25 ductal carcinoma in situ). Nineteen (59%) cancers were confined to the central quadrant of the breast (unifocal). Preoperative imaging (mammography 34 of 34, MRI 13 of 34) detected 15 of 32 (49%) cancers. Mammography detected 11 cancers, accurately demonstrating extent of disease in 9 of 11 patients. MRI detected seven cancers, accurately demonstrating extent of disease in six of seven patients. After positive mammography (n = 11), MRI (n = 5) did not change management. After negative mammography (n = 23), MRI (n = 8) detected otherwise occult disease in 4 of 8 patients, accurately demonstrating extent of disease in 4 of 4 patients and ruling out an underlying cancer in 1 of 1 patient. CONCLUSIONS: Ninety-four percent of patients with biopsy-proved PD as the only physical finding had an underlying cancer and 59% had unifocal disease. Negative preoperative imaging did not reliably exclude an underlying cancer, but the increased sensitivity of MRI detected otherwise occult disease. In the setting of negative mammography, MRI can facilitate treatment planning for patients with PD.