Background: The association of nipple discharge with breast carcinoma has resulted in numerous women undergoing exploratory surgery to exclude malignancy. The aim of this study was to determine whether pre- operative factors can identify those patients that are most at risk of carcinoma. Methods: All patients over a 14- year period ( 1991 - 2005) who had a microdochectomy or subareolar exploration for the evaluation of nipple discharge were assessed. Patient characteristics, pre- operative imaging and pathological findings were analysed. Results: Of the 211 patients included in this study, 116 patients had pathological ( unilateral, uniductal serous or bloody) discharge. On excision, 6% ( n = 7) of patients with pathological discharge and 2.4% ( n = 2) of patients with non- pathological discharge were diagnosed with carcinoma. Overall, major duct excision resulted in the diagnosis of carcinoma in 4.3% ( n = 9), ADH/ LCIS in 4% ( n = 8), papilloma in 39% ( n = 83), and duct ectasia or non- specific benign disease in 53% ( n = 111) of patients. In the patients determined to have malignancy, 44% ( n = 4) were premenopausal. No patient with a non- bloody discharge in the total population analysed ( 28%; n = 59/ 211), or in the population with a pathological discharge ( 21%; n = 24/ 116) was found to have carcinoma upon excision. Conclusion: Microdochectomy or major duct excision performed for nipple discharge resulted in a low rate of malignancy on excision. Conservative management of non- bloody nipple discharge can be considered in patients with no other clinical or radiological signs of malignancy.