Michael Thomas O'Loughlin MD, Presenter: Nothing to Disclose

Tara McLaughlin PhD, Abstract Co-Author: Nothing to Disclose

Ilene Staff PhD, Abstract Co-Author: Nothing to Disclose

Edward Bowen Cronin MD, Abstract Co-Author: Nothing to Disclose

To evaluate the outcomes in patients given a category 3 “probably benign” assessment based on the American College of Radiology MR Breast Imaging Reporting and Data System (BI-RADS) lexicon.

During the course of the study, 473 women consented to outcome analysis at the time of their breast MRI examination. Of these, 158 women (33.4%) were given either a unilateral (104) or bilateral (54) category 3 assessment on their initial study. Sixty women had multiple or bilateral lesions for a total of 226 lesions followed. 31 women with 49 lesions had no follow-up; another 10 lesions (8 women) had incomplete follow-up – not to definitive diagnosis. A total of 167 lesions (in 124 women) were followed to final assessment for a course of up to 32 months after their initial MRI. Outcome endpoints were: Category changed to negative, benign, or suspicious on follow-up studies, positive or negative pathology reports correlating to the finding, multiple stable mammographic and clinical exams after the initial MRI. Patient’s follow-up imaging studies, pathology reports, and clinical reports were evaluated.

Of the lesions given a category 3 assessment on the initial study, 126 were for foci of enhancement, 35 were for benign appearing masses (likely lymph nodes or fibroadenomas), 65 were for non mass-like regions of enhancement. The majority, 162 (97.0%) lesions were benign on final follow-up. Only 5 (3.0%) were eventually determined to be malignant. The malignant lesions on follow up included 4 invasive carcinomas and 1 ductal carcinoma in situ. These malignant lesions were confirmed on average 129 days after the initial category 3 diagnoses.

The MR BIRADS category 3 - probably benign finding, correlated with a high (97%) outcome of benignity on follow up. The majority (76.0%) of the patient’s original assessments were changed on either their first follow-up study (74.3%) or by pathology before the first follow-up exam (1.8%). Of the lesions that were determined to be malignant on follow-up, all remained less than 1 cm in size on final pathology.

The majority of breast MRI lesions assessed as BIRADS category 3 correlated with a benign etiology on eventual follow-up.

O'Loughlin, M, McLaughlin, T, Staff, I, Cronin, E, Prospective Outcome of BIRADS Category 3 "Probably Benign" Breast MRI Lesions.  Radiological Society of North America 2009 Scientific Assembly and Annual Meeting, November 29 - December 4, 2009 ,Chicago IL. http://archive.rsna.org/2009/8002149.html