Does mammary ductoscopy have a role in clinical practice?

BACKGROUND Mammary ductoscopy (MD) is a newly developed endoscopic technique that allows direct visualisation of the mammary ductal epithelium using sub-millimetre fiberoptic microendoscopes inserted through the ductal opening onto the nipple surface. These scopes also provide working channels for insufflation, irrigation, ductal lavage, and possible therapeutic intervention. MD can be performed under local anaesthesia in the office setting. The objective of this study is to assess the technical feasibility of mammary ductoscopy, and examine its role in guiding ductal excision surgery and the early diagnosis of malignancy. METHODS Mammary ductoscopy (MD) was performed using a 1 mm fiberoptic microendoscope (Mastascope TM) in 26 patients (age range: 14-73 years): 13 patients undergoing mastectomy (n = 12) or lumpectomy (n = 1) for ductal carcinoma (including 12 cases of DCIS and one case of infiltrating ductal carcinoma) and 13 patients with pathological nipple discharge (PND) and benign breast imaging and simple discharge cytology. Of the latter group: 10 procedures were performed under local anaesthesia (LA) in the office setting and 3 procedures were carried out under general anaesthesia (GA) to guide duct excision surgery. The ductoscopic appearances in this group were graded between 0 and 5 (D0-D5) according to the degree of suspicion. RESULTS Intraoperative MD was accomplished in 11 (84.6%) of 13 patients undergoing surgery for DCIS. MD was unsuccessful in 2 cases: one patient (aged 73 years) had sclerosis of the nipple and one patient had preoperative vital blue injection in the subareolar region as part of the sentinel node biopsy thus resulting in inadequate visualisation. Intraductal pathology was visualised in 8 (80%) of the 10 cases undergoing mastectomy but ductoscopic cytology was positive for malignancy in only 2 cases (sensitivity = 16%, specificity = 100%). In the office setting, MD was accomplished in 9 (90%) out of 10 patients with PND and was well tolerated (mean pain score = 3.8 out of 10: range 0-7). Of these 10 patients; MD was inadequate (D0) in one patient due to complete occlusion of lumen by the lesion, showed a papilloma in 3 patients (D3), duct ectasia (D2) in 3 patients, irregular thickening of the lumen suspicious of DCIS (D4) in one patient and non-specific benign findings (D2) in 2 patients. Three women with benign ductoscopy and ductoscopy-assisted cytology were reassured and treated conservatively. The remaining 7 patients had ductoscopy-guided duct excision which revealed DCIS in one, papilloma in 4 and benign breast disease in 2 patients. Adequate cellular yield was obtained in 7 (70%) out of 10 cases (benign cytology). The three patients who had MD under GA during microdochectomy had benign endoscopic appearances and final histology (one papilloma and 2 cases of duct ectasia). CONCLUSION MD is technically feasible in most patients and has a potential in the early detection of breast cancer. The procedure can be performed safely in the office setting and should be considered in all patients presenting with a single duct PND. MD has the potential to reduce the number of duct excision procedures and minimise the extent of surgical resection. Ductoscopic cytology is not sufficiently sensitive for the diagnosis of malignancy and the development of a biopsy tool that obtains tissue under direct visualisation is required.