Relationship between BI-RADS and the Results of the Wire-Guided Percutaneous Localization for Non-Palpable Breast Lesions.

Introduction: The aim of this study is to evaluate the relationship between Breast Imaging Reporting and Data System and surgical biopsies that may increase effectiveness of wire-guided percutaneous localizations for non-palpable breast lesions.

Methods: A retrospective review of a sample of 149 patients who underwent wire-guided localization with wide local excision for non-palpable breast lesions at the authors' institution between January 2013 and April 2016. After IRB approval, sample patients' records were reviewed and data were collected concerning their radiological, histological and surgical characteristics.

Results: One (0.67%) complication occurred related to wire migration. There were nine (6.04%) recorded cases of seroma and three (2.01%) cases of hematoma. Breast Imaging Reporting and Data System (BI-RADS) Category 4 was found to have a positive predictive value of 28.4% for breast cancer. Under Category 4 subcategorization 4A, 4B, and 4C, the number of positive lesions were two (6.89%), three (10.34%) and five (17.24%), respectively. Forty (78.43%) of the 51 patients with cancer had negative (i.e., non-cancerous) margins compared to 11 (21.57%) cases that had positive margins after the first procedure.

Conclusion: The BI-RADS Category 4 encompasses the majority of lesions, with approximately 70% of such biopsies lesions later found to be benign. A subcategorization of BI-RADS 4 needs further clarification in distinguishing benign vs malignant imaging characteristics. Future retrospective studies designed to identify benign vs. malignant lesions, confirmed by validating prospective studies, will better clarify a new subcategorization of BI-RADS Category 4, thus allowing surgeons and radiologists to make the best surgical recommendations for their patients.