Atypical ductal hyperplasia diagnosed at stereotactic breast biopsy: improved reliability with 14-gauge, directional, vacuum-assisted biopsy.

PURPOSE: To compare histologic findings of atypical ductal hyperplasia (ADH) at 14-gauge, directional, vacuum-assisted breast biopsy (hereafter, vacuum-assisted biopsy) and at 14-gauge, automated, large-core breast biopsy (hereafter, large-core biopsy) with findings at histologic examination after surgical biopsy. MATERIALS AND METHODS: Nonpalpable breast lesions were diagnosed as ADH at histologic examination after vacuum-assisted biopsy in 88 lesions in seven institutions and after large-core biopsy in 55 previously reported lesions. Histologic findings at subsequent surgical biopsy were compared for the presence of carcinoma. RESULTS: On the basis of histologic findings of carcinoma at surgical biopsy, the diagnosis of ADH was not correct in 26 (48%) of 54 lesions sampled at large-core biopsy and in 13 (18%) of 74 lesions sampled at vacuum-assisted biopsy (Fisher exact test, P < .0004). More tissue specimens were obtained at vacuum-assisted biopsy (mean, 15.8 specimens) than at large-core biopsy (mean, 9.7 specimens). Individual specimens were twice as large at vacuum-assisted biopsy (mean, 34 mg) as at large-core biopsy (mean, 17 mg) (previously reported). CONCLUSION: ADH was diagnosed 2.7 times more reliably at vacuum-assisted biopsy than at large-core biopsy (with no increase in complications) with most of the improvement as a result of acquisition of more than 10 specimens per lesion, but carcinoma was sufficiently underestimated with both methods to necessitate surgical biopsy.