Stereotactic core needle biopsy in non-palpable breast lesions. What number is needed?

Abstract

To investigate whether three stereotactic core needle biopsies (S-CNB) in non-palpable breast lesion are enough for accurate preoperative diagnosis. Between September 1994 and December 2000, 523 patients with mammographically detected breast lesions and who proceeded to surgery were preoperatively stereotactically biopsied with an automated 14-gauge biopsy device. Three samples were taken from each lesion irrespective of whether the lesion presented as "microcalcifications only", "microcalcifications and a mass", or a "mass, architectural distorsion, or stellate lesion without microcalcifications". The histopathology was divided into three subgroups: diagnostic, atypia (ranging from atypical cells to probably cancer), and non-diagnostic material. Post-surgical histopathology diagnosed 454 (87%) malignant tumors and 69 (13%) benign lesions. Three S-CNB correctly diagnosed the malignant tumors in 84% in the subgroup "microcalcifications only". In the category "microcalcifications and a mass", the diagnostic accuracy was 97% and in the subgroup "mass, architectural distorsion, or stellate lesion without microcalcifications" 3 S-CNB resulted in 93% correct diagnostic material. In 19 of the 454 patients (4%) 1, 2 or all 3 preoperative S-CNB showed atypia. In 20 patients (4%), all 3 S-CNB were non-diagnostic. Thirteen of these 20 patients had "microcalcifications only" and 7 had a mass without microcalcifications. Three S-CNB were enough for correct diagnosis in "masses, architectural distorsions, or stellate lesions without microcalcifications" and in "microcalcifications and a mass", but were not sufficient in "microcalcifications only".