The prevalence of carcinoma in palpable vs impalpable, mammographically detected lesions

Concern over excessive numbers of false-positive mammograms, leading to unnecessary investigations and surgical interventions, has been cited as a barrier to mammographic screening for breast cancer. We compared the biopsy results from palpable vs impalpable, mammographically detected lesions from one experienced breast surgeon's practice from July 1980 through July 1989. Overall, there were 372 biopsies in 346 women. Of 143 biopsies for palpable abnormalities, 48 (34%) yielded a primary malignant lesion. The length of the palpable cancers averaged 3.7 cm (median, 2.8 cm). Sixteen (33%) of the 48 biopsies were in patients who had positive axillary lymph nodes, and five (10%) were in patients who had distant metastases at the time of biopsy. Of 229 biopsies for impalpable, mammographically detected lesions, 72 (31%) yielded a primary breast carcinoma. Excluding 34 carcinomas that had only calcifications, the length of the mammographically detected tumors averaged 2.0 cm (median, 1.5 cm). Eleven (15%) of the 72 biopsies were in patients who had positive axillary nodes, and none were in patients who had distant metastases at the time of biopsy. The positive predictive values (number of cancers detected divided by the number of biopsies recommended) were not significantly different when comparing biopsies indicated for palpable, clinically detected (34%) vs impalpable, mammographically detected (31%) abnormalities (p = .669). However, the mammographically detected cancers were smaller, more often noninvasive (32% vs 4%), less often associated with axillary metastases (15% vs 33%), and without distant metastases (0% vs 10%).