P2-08-11: Correlation of Mammography, Ultrasound and MRI in Patients with Nipple Discharge.

Abstract

Abstract BACKGROUND: Breast cancer mainly originates in the epithelial lining of milk ducts or lobules. Once early pre-malignant or malignant changes occur in the ductal lining there is a long latent period before they manifest symptoms or are detected on imaging. Symptomatic nipple discharge (SND) is a common symptom noted in approximately 5% of patients. Although, usually benign it may be the earliest sign of cancer. The current standard evaluation of patients with nipple discharge includes clinical examination (CE), mammography (MMG) and ultrasound (US). Contrast enhanced MRI is considered the most sensitive imaging modality for the breast and reports suggest it may be of benefit in the evaluation of nipple discharge. We have combined MRI with standard evaluation in an ongoing prospective study to determine its utility in patients with SND. PATIENTS AND METHODS: Between July 2004 and 31st May 2011, 75 women underwent surgery for SND. All were initially evaluated by CE, MMG and US following which a contrast enhanced breast MRI was requested. MRI was performed on a 1.5T MR unit using a dedicated breast coil. A STIR sequence in the coronal and sagittal planes along with an unenhanced fat-suppressed T1 W sequence was used to evaluate the ducts. The imaging findings were used to direct surgical excision. All specimens were oriented and evaluated by the pathologist by sectioning at 1 cm interval, from the ductal end to the periphery. The results of preoperative MMG, US and MRI findings were compared with pathology. RESULTS: The incidence of cancer in this series was 29% and high risk lesions were noted in 43%. MMG and US were performed in 75 patients. MRI was requested in all 75 patients but could be performed in 64. MRI proved to be the most sensitive technique for detection of an abnormality in the breast (92%) compared to MMG 25%, and US 63%. MMG, US and MRI were suggestive of benign change in 64 (85%), 65 (87%) and 43 (67%) patients respectively, of which 14 (22%), 15 (23%) and 7 (16%) were malignant on pathology. MMG, US and MRI reported suspicious change in 11(15%), 10 (13%) and 21(33%) patients respectively of which 8 (73%), 7 (70%) and 9 (43%) patients had cancer on pathology. MRI had the highest sensitivity for malignancy (54%) compared to MMG (36%) and US (32%) but MRI had a low specificity (74%) (MMG 94%; US 94%). However, MRI had a high negative predictive value (86%) (MMG 78%; US 73%). Of the 22 patients with cancers, 12 were deemed to be normal on CE, MMG and US examination. Thus standard evaluation would have missed 54% of the cancers detected in this series. CONCLUSIONS: Nipple discharge is associated with significant incidence of cancer and high risk changes. The majority of these are missed on standard evaluation (CE, MMG, US). MRI is the most sensitive method for the detection of an abnormality in these patients. It defines the location and extent of the abnormality and can guide surgical excision. However, MRI is unable to accurately differentiate malignant from benign lesions. Hence, excision of an identified abnormality and pathological evaluation remains essential. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P2-08-11.