During a 3-year period (1987)(1988)(1989), 60 fine-needle aspiration biopsies (FNAs) were obtained from new breast lesions in patients previously treated by radiation and surgery for breast carcinoma.
The lesions occurred at or near the site ofprevious excision, 3-11 7 months after initiation of radiotherapy. FNAs were classijied as follows: acellular (11); negative (29); atypical (13); suspicious (4); and positive (3). For statistical analysis, acellular, negative, and atypical diagnoses were considered negative jndings, and suspicious and positive diagnoses were considered positive jindings. On the basis of subsequent biopsy and/or patient follow-up, FNA yielded a sensitivity of 86%, a specificity of 98%, apositivepredictive value of 86%, a negative predictive value of 98%, and an eficiency of 97%. Excluding cystic lesions, the most reliable criterion for distinguishing malignant from benign lesions was the abundance of epithelial cells, both singly and in large clusters. Cellular characteristics were less helpful, since nuclear atypia was seen in both benign and malignant lesions. I t is concluded that ( I ) FNA is a reliable technique in the evaluation of the irradiated breast; (2) when performed by an experienced operator, an acellular aspirate may be interpreted as evidence against recurrent carcinoma; and (3) epithelial atypia must be interpreted with caution to avoid a false-positive diagnosis.