Subclassification of benign breast disease by fine needle aspiration cytology. Comparison of cytologic and histologic findings in 265 palpable breast masses.

The cytologic and histologic features of 265 benign breast masses were analyzed in order to examine the ability of fine needle aspiration cytology to accurately subclassify benign breast lesions. Two hundred two of the masses were pure histologic examples of benign breast lesions (72 nonproliferative fibrocystic change, 27 proliferative fibrocystic change, 65 fibroadenoma, 12 abscess, 8 fat necrosis, 7 papilloma, 7 duct ectasia, 2 tubular adenoma, 1 sclerosing adenosis, 1 microglandular adenosis), and 63 masses were mixed lesions. Part I of the study consisted of retrospective comparison of the original cytologic diagnoses with the histologic diagnoses. A nonspecific descriptive diagnosis had been rendered in 135 of 265 (51%) cases, and these descriptive diagnoses corresponded to fibrocystic change in the majority of cases (70%). A specific benign cytologic diagnosis had been made in 130 of 265 (49%) cases, and overall the specific diagnosis was correct in 80% of cases. Part II of the study consisted of the semiquantitative scoring of the cytologic findings of the 202 pure examples of benign breast masses and statistical analysis of differences in the expression of cytologic features between the different types of lesions. Overall cellularity, amount of bipolar stripped nuclei, amount and architectural arrangement of epithelium, epithelial atypia/pleomorphism/nuclear overlapping and amount of apocrine metaplasia, foam cells and stroma were the cytologic parameters that were statistically significant (P < .05) in distinguishing between the cases of fibroadenoma, abscess, papilloma, fat necrosis, duct ectasia and fibrocystic change as a group. No cytologic parameter reached statistical significance in distinguishing between proliferative and nonproliferative fibrocystic change. We conclude that the majority of benign breast lesions yield characteristic cytologic findings that allow their subclassification when sufficiently sampled by fine needle aspiration. The distinction between proliferative and nonproliferative fibrocystic change is less reliable, and cytologic differences observed within this spectrum did not reach statistical significance.