Nipple aspirate fluid (NAF) cytology is a simple noninvasive method to study cells exfoliated into the ductal system of the breast. In the current study, the significance of cytologic findings in NAF was determined by correlating them with histopathologic findings from corresponding breast tissue. Cytologic-histologic correlations of NAF were performed in only a few studies.
Nipple aspirate fluid was collected by breast massaging and by using a breast aspiration device from 74 women with biopsy confirmed intraductal or invasive carcinoma with or without a history of preoperative neoadjuvant chemotherapy. Cytospin preparations were Pap stained. The number of epithelial cells was quantitated and foamy macrophages were semiquantitatively scored. Cytologic findings were categorized as insufficient for diagnosis (less than 10 epithelial cells), benign, mild atypia, marked atypia or suspicious, and malignant. Finally, they were correlated with tissue findings.
Nipple aspirate fluid was obtained from 74 women, including 24 who had received preoperative neoadjuvant chemotherapy. The median age of patients was 54 years. A mean volume of 57 microL NAF and a mean of 149 epithelial cells were obtained. Foamy macrophages were present in 51 (70%) of the specimens. There was a significant correlation between the presence of epithelial cells and foamy macrophages (P < 0.001). Patients treated with chemotherapy had fewer epithelial cells in their NAF compared with patients who were not treated with chemotherapy. Thirty specimens (41%) were inadequate for diagnosis, 34 were (46%) benign, 5 (7%) were mildly atypical, 1 (1%) was markedly atypical, and 4 (5%) were malignant. Of the five cases with mildly atypical cytology, three were intraductal papilloma, one was low-grade papillary intraductal carcinoma, and one was low-grade intracystic papillary carcinoma with invasion in the corresponding tissue specimen. The single case with markedly atypical NAF cytology had extensive ductal carcinoma in situ (DCIS). Of the four cases with malignant NAF cytology, two were extensive DCIS and two had invasive carcinoma with extensive DCIS in the breast specimen. Overall, 3 (27%) of 11 cases of DCIS were detected in NAF and only 2 (4%) of 52 invasive carcinomas including the only two cases with extensive DCIS were detected in NAF.
The probability of detecting malignant cells in NAF is dependent on the extent of DCIS and nipple involvement by DCIS. Nipple aspirate fluid is not a sensitive test for detecting invasive carcinoma of the breast. Atypical cytology in NAF is associated with papillary lesions in the underlying breast.