Time-Related Changes in Yield and Harms of Screening Breast Magnetic Resonance Imaging

Abstract Purpose Breast magnetic resonance imaging (MRI) is accepted as a useful adjunct to screening mammography for women at high risk for breast cancer. Nevertheless, concerns about false-positive findings remain, and data about MRI harms and yields are limited. The aim of this study was to quantify harms and yields of breast MRI over time in a large series of patients. Methods A retrospective review was performed of patients at increased risk for breast cancer who underwent annual screening digital mammography and MRI from 2007 to 2013. Harms were defined as events not producing a breast cancer diagnosis (ultrasonography [US], imaging-guided core or surgical biopsy procedure, recommendation for short-term follow-up, or a combination). Results Of 350 high-risk patients offered MRI screening, 320 underwent 757 screening MRI procedures over time. The median age at the first MRI was 48 years. All patients met American Cancer Society criteria for annual screening breast MRI. Total harms were highest with the first MRI procedure and decreased with subsequent MRI screening. Of 75 biopsy procedures performed, including 58 US- or MRI-guided core biopsy procedures and 17 surgical biopsy procedures, 6 specimens were found to be malignant, including 2 resulting from biopsy procedures performed based on findings from the first MRI scan, 0 from the second MRI scan, 3 from the third MRI scan, and 1 from the fourth MRI scan. Conclusion Among women followed with screening MRI, the number of harms was shown to decrease over time. Breast cancer continued to be detected in MRI studies performed over time. This study demonstrates the utility of MRI screening performed over time in high-risk women.