American Journal of Roentgenology (AJR)

AJR InBrief Banner 

AJR October 2018

Bethany Neill
Massachusetts General Hospital
Boston, MA

“Utility of Breast MRI for Further Evaluation of Equivocal Findings on Digital Breast Tomosynthesis”

This study in the October issue of AJR evaluated the utility of MRI as a problem-solving tool for equivocal findings on diagnostic digital mammography (DM) and digital breast tomosynthesis (DBT). As DBT becomes increasingly widespread, breast imaging radiologists need an algorithm to address the small number of inconclusive findings that remain equivocal despite thorough DM-DBT and ultrasound. Breast MRI is a useful adjunctive tool in these selected cases.

Breast MRI performed from March 2011 to November 2014 were retrospectively reviewed to identify examinations performed to further assess equivocal findings on combination DM-DBT. All patients underwent diagnostic ultrasound in conjunction with their DM-DBT examination. Imaging reports were retrospectively reviewed for BI-RADS findings and assessments of the diagnostic DM-DBT and diagnostic MRI. A review of the electronic medical record provided demographic data, cancer diagnoses, and pathology results. Differences in positive predictive value and negative predictive value were compared for DM-DBT and MRI using a generalized estimating equation for correlated binary data.

Of 5330 MRI examinations performed during the study period, 67 (1.3%) were for evaluation of an equivocal finding, including 27 (40%) asymmetries, 16 (24%) focal asymmetries, five (8%) masses, and 19 (28%) architectural distortion. MRI correlates were identified in 22 of 67 (33%). Biopsies yielded cancer in five of 67 (7%). For MRI, the positive predictive value and negative predictive value were 19% and 98%, respectively, compared with 6% and 90% on DM-DBT (p = 0.009 and 0.059, respectively). The frequency of breast MRI recommendations to evaluate equivocal findings exponentially decreased in the 3 years following DBT implementation.

In this Q&A, Neill discusses the authors’ ideas behind the study and the outlook for further research.

What is the background for this study? How did you and your co-authors come up with the idea?

Neill: With increasingly widespread implementation of DBT in the United States, radiologists will need an algorithm to address the tiny number of inconclusive findings that remain equivocal despite thorough evaluation with diagnostic DBT and ultrasound. The utility of MRI for further evaluation of an equivocal finding on digital breast DBT examinations had not been previously investigated.

What should readers take away from your article?

Neill: Three things. First, equivocal findings on DBT are infrequent, so breast radiologists seldom need to recommend breast MRI for further evaluation. With increasing DBT experience, radiologists’ breast MRI recommendations for equivocal DBT findings should decrease.

Second, because breast MRI has a high negative predictive value, a negative MRI may be reassuring to the patient and radiologist that short term follow-up diagnostic DBT would be a reasonable next step.

Third, radiologists should be careful of one-view findings on DBT, similar to mammography, because of the small frequency of malignancy.

What recommendations do you have for future research as a result of this article?

Neill: Our study investigated equivocal DBT findings at a single institution, so our sample size is limited. We hope to see our findings confirmed at other institutions.


Do you have a story idea for InBrief?
Contact Mike Mason, Communications Manager, at for more information.