Does core needle breast biopsy accurately reflect breast pathology?

Abstract

Core needle breast biopsy (CB) has replaced excisional biopsy as the initial diagnostic biopsy procedure for many suspicious breast lesions; however, CB remains a sampling procedure. The purpose of this study was to determine the degree of agreement between histology obtained at CB and that obtained at a subsequent excisional procedure (EP). We hypothesized a high degree of agreement. Data were collected prospectively for 3035 CBs performed by breast radiologists using either ultrasound or stereotactic guidance between January 1995 and July 2002, 1410 (46%) of which had a subsequent EP within 1 year. Histologic categories were defined as invasive breast cancer, duct carcinoma in-situ, atypia/lobular carcinoma in-situ, and benign. The principal histology (PH) from CB and EP was identified and compared. Overall, there was moderate agreement (kappa=0.669) between CB and EP histology. Complete agreement occurred in 1168 (83%) procedures. For the remaining 242, the PH was identified only at CB for 78 (5%) procedures, and only after EP for 164 (12%) procedures. Although the majority (83%) of CB and EP demonstrated exact histologic agreement, CB was diagnostic for 1246 (88%) procedures.