False‐negative core needle biopsies of the breast

We agree with Shah et al. that many false-negative core needle biopsy results can be avoided by standard “post-core needle biopsy” protocols that include systematic evaluation of the correlation between the mammographic appearance of the lesion and the histologic diagnosis.1 However, not all false-negative results can be prevented. In a multiinstitutional study in the Netherlands (including patients with nonpalpable breast lesions) that was performed between 1997–2000 in 19 hospitals, 822 of 871 stereotactically guided, 14-gauge, core needle biopsies (94%) were confirmed surgically according to a standard protocol.2 A total of 290 core needle biopsies were found to be invasive carcinoma, 190 were found to be ductal carcinoma in situ (DCIS), 26 were found to be high risk (atypical hyperplasia or lobular carcinoma in situ), 30 were normal breast tissue, and 322 were benign. False-negative results were defined as a benign core needle biopsy diagnosis followed by a malignant diagnosis (i.e., DCIS or invasive carcinoma). A panel comprised of two expert breast radiologists and three expert breast pathologists reviewed all radiologic and histologic material to elucidate reasons for missing the carcinoma diagnosis. Of the total 322 benign large core needle biopsy diagnoses, 13 (4%) appeared to be a false-negative result. In eight of these cases, the reasons for missing the carcinoma diagnosis were considered as “avoidable” (e.g., microcalcifications were not sampled [two cases], digital images showed that the biopsy needle had not penetrated the mass lesion [one case], or high-risk lesions [three cases] or in situ carcinomas [two cases] in the core needle biopsy were not recognized by routine pathology). However, in five cases (four cases of DCIS and one case of invasive carcinoma), no “avoidable” reason for missing the carcinoma diagnosis was found. In all these cases, histologic findings were found to be in perfect correlation with the mammographic abnormalities and the biopsy procedures were technically “correct” (i.e., the biopsy needle was penetrating the lesion on digital images, at least 50% of the core needle biopsies contained microcalcifications). In three of the five cases, mammographic abnormalities were caused by benign breast disease, whereas malignancy in the direct vicinity of the benign calcifications was not (yet) visible mammographically and was not sampled by core needle biopsy. In the two other cases, the reasons for missing the carcinoma diagnosis were not understood. In all five cases malignancy was detected by open diagnostic breast surgery, which was standard protocol during the study. However, these tumors would have been missed by large core needle biopsy, even when standard protocols were used as proposed by Shah et al.1 Biopsy techniques that obtain larger amounts of tissue, such as the 11-gauge, vacuum-assisted core needle biopsy, may reduce “unavoidable” missed cases.3, 4 We want to emphasize the fact that false-negative core needle biopsies can never be completely avoided. However, in our experience, the proportion of missed carcinomas is low; in our series, 4% of cases were missed (13 of 322 cases) and only 1.5% (5 of 322 cases) were considered “unavoidable.” We therefore agree that large core needle biopsy should be considered to be a reliable alternative to open biopsy, provided that a standard protocol, comprising mammographic-histologic correlation, is followed. Nevertheless, physicians and patients should keep in mind that despite all precautions, unavoidable false-negative results do occur, although very rarely.