Objective: The objective of this study was to compare the underestimation rate of invasive carcinoma in cases with ductal carcinoma in situ (DCIS) at percutaneous ultrasound-guided core biopsies of breast lesions between 14-gauge automated coreneedle biopsy (ACNB) and 8-or 11-gauge vacuum-assisted biopsy (VAB), and to determine the relationship between the lesion type (mass or microcalcification on radiological findings) and the DCIS underestimation rate. Methods: We retrospectively reviewed imaging-guided biopsies of breast lesions performed from February 2003 to August 2008. 194 lesions were diagnosed as DCIS at ultrasound-guided core biopsy: 138 lesions in 132 patients by 14-gauge ACNB, and 56 lesions in 56 patients by 8-or 11-gauge VAB. The histological results of the core biopsy samples were correlated with surgical specimens. The clinical and radiological findings were also reviewed. The histological DCIS underestimation rates were compared between the two groups and were analysed for differences according to the clinical and radiological characteristics of the lesions. Results: The DCIS underestimation rate was 47.8% (66/138) for 14-gauge ACNB and 16.1% (9/56) for VAB (p,0.001). According to the lesion type on sonography, DCIS underestimation was 43.4% (63/145) in masses (47.6% using ACNB and 15.8% using VAB; p50.012) and 24.5% (12/49) in microcalcifications (50.0% using ACNB and 16.2% using VAB; p50.047).
Conclusion:The underestimation rate of invasive carcinoma in cases with DCIS at ultrasound-guided core biopsies was significantly higher for ACNB than for VAB. Furthermore, this difference does not change according to the lesion type on ultrasound. Therefore, ultrasound-guided VAB can be a useful method for the diagnosis of DCIS lesions presented as either mass or microcalcification.