Characteristics and Outcomes of BI-RADS 3 Lesions on Breast MRI

Background: There are few data regarding the use and outcomes of Breast Imaging Reporting and Data System (BI-RADS) 3 assessment on breast magnetic resonance imaging (MRI). The aim of this study was to describe the imaging findings prompting a BI-RADS 3 assessment and to report their outcomes, including the timing of follow-up examinations. Patients and Methods: We performed a retrospective study evaluating 199 breast lesions in 186 patients who were assigned a BI-RADS 3 assessment on breast MRI over a 5-year period. Clinical and imaging features were recorded. For outcomes analysis, lesions were considered benign if they showed 2 years of MRI stability, if they were declared benign during follow-up, or if the patient underwent biopsy with benign pathology results. Clinical and imaging features of BI-RADS 3 lesions associated with malignancy were assessed by the Fisher exact test, with P <.05 considered significant. Results: Of the 199 breast MRI lesions assigned a BI-RADS 3 assessment, 80 (40%) of 199 were non–mass enhancement, 61 (31%) were a single focus, and 58 (29%) were masses. A total of 131 lesions (66%) were eligible for outcome analysis after excluding those lost to follow-up; 4 (3%) were diagnosed as malignant within the 2-year follow-up. Masses assigned a BI-RADS 3 assessment were more likely to be malignant during follow-up than non–mass enhancement or single focus (P <.05). Conclusion: Despite limited data on the use of BI-RADS 3 at breast MRI, there is a low malignancy rate of 3% at our institution. Additional studies are needed to further define the appropriate use of BI-RADS 3 on breast MRI.