This study was designed to investigate the accuracy of contrast-enhanced MR imaging in the locoregional staging of symptomatic primary breast cancer and to determine the impact of contrast-enhanced MR imaging in planning surgical management. Ninety patients with primary breast cancer (including two bilateral cancers) diagnosed and treated on the basis of conventional triple assessment (clinical, cytologic, and mammographic examination) underwent MR imaging at 1.0 T using a three-dimensional fast low-angle shot T1-weighted pulse sequence before and after contrast enhancement. A short inversion time inversion recovery sequence was also obtained to evaluate the axilla of each patient. After resection, tumors were histopathologically mapped in detail and correlated with the extent of contrast enhancement on MR imaging. On the basis of triple assessment, 53 cancers were treated by wide local excision, of which 17 (32%) had positive margins at excision. Residual disease at reexcision was detected in eight of these 17 patients, a finding that correlated accurately with the extent of contrast enhancement on MR imaging. MR imaging was more accurate than mammography in determining invasive tumor size (r2 = .93 versus r2 = .59), in depicting multifocality and extensive intraductal component (sensitivity, 81% versus 62%), and in assessing nipple-retroareolar complex. MR imaging-histopathologic correlation was possible in 75 axillae. Sensitivity and specificity for axillary node metastases were 90% and 82%, respectively. MR imaging of the breast has value in the preoperative locoregional staging of symptomatic primary breast cancer and is useful in planning a single definitive surgical resection in patients with breast cancer.