Nicole Nakyung Lee MD, Presenter: Nothing to Disclose

Tracy Frazee, Abstract Co-Author: Nothing to Disclose

Zhongze Li, Abstract Co-Author: Nothing to Disclose

Steven P. Poplack MD, Abstract Co-Author: Research Grant, Hologic, Inc

To determine if 2D (FFDM) screening recall rates have been impacted by the clinical implementation of 2D/3D digital breast tomosynthesis (DBT).

Our institution initiated use of 2D/3D DBT as a screening modality in February 2012 in one of three screening mammography units. A retrospective review of screening mammography was performed for a two-year period before and after the introduction of tomosynthesis. Aggregate FFDM recall rates from February 2010-January 2012 were compared to aggregate FFDM recall rates from February 2012-January 2014 using a Chi-square test. Both sets of recall rates (before vs after DBT) were stratified by density and a three dimensional analysis was performed using a Cochran-Mantel-Haenszel test.

A total of 24,384 FFDM screening mammograms were performed in the two years before implementation of 2D/3D DBT; 17,210 FFDM screening mammograms were performed in the two years after the addition of 2D/3D DBT to the screening practice. The aggregate FFDM recall rate was significantly lower after implementation of 2D/3D DBT, decreasing from 8.75% to 7.76% (p=0.0003). When stratified by density, recall rates for each time period (before vs after, respectively), were: extremely dense 9.41% vs 9.47%; heterogeneously dense 10.64% vs 9.25%; scattered 8.49% vs 7.71%; fatty 4.67% vs 4.96%. The p-value from the Cochran-Mantel-Haenszel test was 0.008.  

The recall rate of FFDM may be positively influenced by the partial implementation of 2D/3D DBT in screening mammography. The etiology of this effect is unclear, but may be related to a learning curve from experience with 2D/3D DBT. 

Addition of 2D/3D DBT to clinical screening practice may indirectly result in a reduction of 2D (FFDM) screening recall rates, which would contribute further to the beneficial effect of DBT implementation in a breast cancer screening program.  

Lee, N, Frazee, T, Li, Z, Poplack, S, A Comparison of FFDM Screening Recall Rates Before and After the Addition of 2D/3D Digital Breast Tomosynthesis: Is there a “trickle-down” Effect?.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14045613.html