Can we see what is invisible? The role of MRI in the evaluation and management of patients with pathological nipple discharge.

INTRODUCTION: The aim of this study was to determine the ability of MRI to identify and assess the extent of disease in patients with pathological nipple discharge (PND) with an occult malignancy not evident on standard pre-operative evaluation with mammography and ultrasound. METHODS: Patients presenting to the breast unit of Imperial College Healthcare NHS Trust between December 2009 and December 2018 with PND and normal imaging were enrolled in the study. Pre-operative bilateral breast MRI was performed in all patients as part of our protocol and all patients were offered diagnostic microdochectomy. RESULTS: A total of 82 patients fulfilled our selection criteria and were enrolled in our study. The presence of an intraductal papilloma (IDP) was identified as the cause of PND in 38 patients (46.3%), 14 patients had duct ectasia (DE-17%) and 5 patients had both an IDP and DE. Other benign causes were identified in 11 patients (13.4%). Despite normal mammography and ultrasound a malignancy was identified in 14 patients (17%). Eleven patients had DCIS (13.4%), two had invasive lobular carcinoma and one patient had an invasive ductal carcinoma. The sensitivity of MRI in detecting an occult malignancy was 85.71% and the specificity was 98.53%. The positive predictive value was 92.31% and the negative predictive value was 97.1%. CONCLUSIONS: Although a negative MRI does not exclude the presence of an occult malignancy the high sensitivity and specificity of this diagnostic modality can guide the surgeon and alter the management of patients with PND.

Surveillance of probably benign (BI-RADS 3) lesions in mammography: what is the right follow-up protocol?

BI-RADS-3 is a category in mammography for probably benign lesions and for which periodic follow-up with repeat imaging is recommended. At our institution repeated mammograms are performed at 6, 12, 18, and 24 months. The purpose of this study was to assess the significance of 18-month mammogram for evaluation of BI-RADS-3 lesions. Following IRB approval, electronic medical records and picture archiving and communications system were used to review 121,862 consecutive mammograms between February, 2002-May, 2009. A total of 8,400 patients with BI-RADS-3 mammograms were identified. Of these, 7,632 patients were followed until completion of 24 month mammogram or biopsy following an upgrade in their BI-RADS status. Over the follow-up, 197 patients received an upgrade in their BI-RADS status of which 179 were biopsied. Histopathologic results were reviewed. The majority of the BI-RADS-3 lesions were upgraded at 6-month mammogram (n = 150, 76.1%) followed by 32 (16.2%), 11 (5.6%), and 4 (2.0%) at 12, 18, and 24 month mammograms respectively. Thirty-four of 179 upgraded and biopsied lesions were found to be malignant. From these 27 (79.4%), 3 (8.8%), 3 (8.8%), and 1 (2.9%) lesions were identified at 6, 12, 18, and 24 month mammograms respectively. At the 18-month mammogram 3/7,632 lesions (0.04%) were found to be malignant. The vast majority of malignant lesions (88.2%) were detected within the first 12-months of follow-up. Only three of 179 biopsies (1.7%) were malignant at 18-month follow-up. Based on those results a 6-, 12-, and 24-month follow-up protocol for BI-RADS-3 lesions is sufficient.