Rate of diagnostic underestimation in ultrasound-and mammography guided core needle biopsies of breast lesions

Resumo

Introduction: Core needle biopsy (CNB) of suspicious breast lesions classified by the Breast Imaging Reporting and Data System (BI-RADS^®) can be guided by ultrasound (US) or mammography (MMG), and it is important to monitor the rate of diagnostic underestimation when performing these procedures. Methods: This is a retrospective study of 524 breast lesions (BI-RADS^®≥3) submitted to core needle biopsy (CNB) at a tertiary hospital in Curitiba/PR between 2019 and 2022. Of these, 261 were subsequently submitted to surgical excision. The underestimation frequency was evaluated at two time points: comparing BI-RADS^® with CNB histology and with surgical excision; and between CNB histology and the surgical excision report, separately for each guiding method – Ultrasound-Guided Core Needle Biopsy (US-CNB) and Mammography-Guided Core Needle Biopsy (MMG-CNB). Results: A total of 83 MMG-CNB and 441 US-CNB were performed. The BI-RADS^® 5 malignancy rate in US-CNB was 89.7% in percutaneous biopsy and 92.7% in surgical excision (p<0.05). When comparing histology from MMG-CNB and surgical excision, underestimation was significantly higher in MMG-CNB, with 7/29 lesions (24.13%) underestimated versus 9/232 lesions (3.89%) in US-CNB (p<0.05). Conclusions: There was no underestimation between the BI-RADS^® classification and the histology of CNB. In the latter, however, underestimation was observed in comparison with surgical excision, especially in MMG-CNB, indicating the need for a better evaluation of the lesions to avoid diagnostic errors and ensure greater accuracy in clinical treatment.