A prospective study of double diagnosis of nonpalpable lesions of the breast.

Approximately three-fourths of open biopsies of the breast performed for mammographically detected suspicious lesions are shown histologically to be benign. Under the narrow conditions described herein, stereotaxic fine-needle aspiration (FNA) can identify these lesions with an accuracy of more than 90 per cent and a false-negative rate of 5 per cent. In an effort to reduce this failure rate, the mammographic appearance and stereotaxic FNA results of these lesions each were given scores on a scale of zero (benign) to five (malignant), to derive an over-all risk score prospectively applied to 264 suspicious occult lesions of the breast prior to open, biopsy. While all 264 lesions could be assigned a mammographic score, adequate tissue for assignment of a cytologic score could be obtained from 150 lesions. Of the 150 evaluable lesions, 53 were malignant and 97 were benign, historically. With a total score of two as the threshold for open biopsy, 21 of 150 (14 per cent) were proved to be benign, with no false-negative findings. If the total threshold score mandating an open biopsy was raised to four, the comparable figures were 61 of 150 (40 per cent) benign lesions and two false-negative instances of carcinoma in situ. Provided adequate tissue is aspirated for cytologic examination, we conclude that this algorithm has practical value in the management of nonpalpable lesions of the breast in that it can reliably identify a fraction of the benign lesions and spare these patients an operation.