Objective: Image-guided breast biopsy is commonly used for diagnosis of breast lesions. Breast ultrasound (US), which is also used for screening purposes, is an important tool to guide biopsies. In this study, we evaluated the radiologic-histopathologic correlation in patients who underwent US-guided breast biopsy.
Material and Methods: A total of 126 biopsies from 116 consecutive cases were included. Patients US and histopathological findings were retrospectively reviewed.
Results: All patients were female. Median age was 44±12 (range; 16-66 years old). Two patients (2%) had bilateral, 8 (7%) had multifocal lesions. Breast Imaging Reporting and Data System (BI-RADS) was used for 115 lesions (91%). Three cases (2%) were BI-RADS 2, 27% (n=34) BI-RADS 3, 35% (n=44) BI-RADS 4, 25% (n=32) BI-RADS 5 and 2% (n=2) BI-RADS 6. Eight biopsies composed of normal breast tissue, which had been scored as BI-RADS 3 or 4, were considered inadequate. More than one-third (37%; n=47) were malignant as 28% (n=35) were consistent with fibroepithelial lesions and 11% (n=14) with inflammatory lesions. Major radiologic-histopathologic discordance was observed in only 2 cases, while there was minor discordance in 14. Ten of the 14 cases (11%) with minor discordance were BI-RADS 4 lesions and minor discordance was more common for benign lesions (p=0.013).
Conclusion: Our findings demonstrate a high radiologic-histopathologic correlation rate in US-guided breast biopsy samples. We observed the highest discordance in BI-RADS 4 lesions, suggesting that histopathological verification is necessary in patients with BI-RADS 4 lesions to exclude malignancy.