Changes in the Utilization of the BI-RADS Category 3 Assessment in Recalled Patients Before and After the Implementation of Screening Digital Breast Tomosynthesis

Abstract

The purpose of this study was to compare the utilization of the Breast Imaging Reporting and Data System (BI-RADS) category 3 assessment in patients recalled from screening before and after the implementation of digital breast tomosynthesis (DBT). This was a retrospective review of 11,478 digital mammography (DM) screening exams and 9350 DM+DBT screening exams. Lesions assigned a BI-RADS category 3 at diagnostic exam were classified as architectural distortions, asymmetries, calcifications, masses, and "other" and followed for a minimum of 2 years. The addition of DBT to screening DM resulted in a 30.4% relative reduction (10.3 women per 1000) in the utilization of BI-RADS category 3 compared to screening DM alone (3.4% for DM versus 2.4% for DM+DBT; p < 0.0001). There was a statistically significant change in the distribution of category 3 findings with DM+DBT characterized by an increase in calcifications and architectural distortions and a decrease in asymmetries. There was no change in category 3 assessment for masses. Although both cohorts had delayed cancer detection rates that exceeded the recommended 2% benchmark (2.3% for DM and 3.6% for DM+DBT), when limited to invasive malignancies, the delayed cancer detection rates were below the 2% benchmark (1.5% for DM and 0.9% for DM+DBT). Screening DM+DBT resulted in a 9.2% relative reduction in recall rate compared to DM (13.0% for DM versus 11.8% for DM+ DBT, p = 0.012). Implementation of DBT in the screening population decreased the overall number of patients assigned to short-term follow-up by 10.3 per 1000 women while maintaining comparable rates of delayed cancer detection.