Objective: to evaluate the diagnostic accuracy of ultrasound-guided core needle biopsy of breast tumors.
Methods: six hundred and sixty-seven cases of core needle biopsy of breast encountered during the period from January, 2004 to June, 2007 were retrieved from the archival file and retrospectively reviewed. The core needle biopsy diagnoses were correlated with the histologic findings of the subsequent surgical excision specimens. The discrepancies were further analyzed.
Results: three hundred and eighty-two patients had core needle biopsy diagnosis followed by local excision, breast conservation surgery or mastectomy. Two hundred and eighty-one cases were confirmed to have malignancy in the surgical specimens. Review of the corresponding core needle biopsies showed 4 false-negative cases, no false-positive cases, 28 cases with underestimation and 2 cases with overestimation. The false-negative rate was 1.4% (4/281). The rate of underestimation for ductal carcinoma-in-situ was 6/11. The diagnostic accuracy of core needle biopsy was 94.7% (266/281).
Conclusion: in order to improve the diagnostic accuracy of core needle biopsy of breast tumors, recognition of the limitation of the procedure, application of immunohistochemistry and awareness of potentially rare entities are important.