Objective: The purpose of this study was to ascertain whether a comparison of preoperative mammograms with radiographs of biopsy specimens is useful for determining the presence of residual breast tumor after an initial excision of impalpable breast cancer.
Materials and methods: Radiographs of tissue specimens obtained at the initial biopsy of 125 impalpable breast cancers were compared with the preoperative mammograms to determine if the lesion seen on the mammogram was completely excised. All tumors were impalpable, necessitating preoperative wire localization. The biopsies were usually excisional, and no special efforts were made to remove additional surrounding tissue. Specimen radiographs were rated as showing no residual tumor if the lesion appeared to be completely excised, showing residual tumor if the lesion did not appear to be completely removed, or indeterminate for residual tumor if adequacy of excision was uncertain. The presence or absence of tumor at the margins of the surgical specimen was determined by histologic examination. These results were correlated with the presence or absence of residual breast cancer at subsequent mastectomy (n = 71) or reexcision (n = 54).
Results: The specimen radiograph showed complete excision of the mammographic lesion in 79 (63%) of the 125 cases. Tumor was found at mastectomy or reexcision in 35 of these cases, giving a false-negative rate for residual tumor of 44%. Incomplete excision was shown on 39 specimen radiographs. Eight of these 39 had no residual tumor, giving a false-positive rate of 21%. The sensitivity of the specimen radiograph for predicting the presence of residual tumor was 49%, specificity was 77%, and overall accuracy was 62%. The accuracy of pathologic examination of the margins of the biopsy specimen for predicting residual breast cancer was 58%. The specimen radiograph alone correctly identified 18 cases of residual tumor in which biopsy margins were indeterminate or negative for the presence of tumor.
Conclusion: The specimen radiograph is not reliable enough to be used alone for determining the presence or absence of residual breast cancer after the initial excision of impalpable breast cancer. However, it can be of value in predicting the presence of residual tumor in those cases in which results of pathologic examination of biopsy specimen margins are either indeterminate or negative for the presence of tumor but the specimen radiograph shows incomplete excision of the mammographic lesion.