Kyung Jin Nam MD, Abstract Co-Author: Nothing to Disclose

Boo-Kyung Han MD, PhD, Presenter: Nothing to Disclose

Eun Sook Ko MD, Abstract Co-Author: Nothing to Disclose

Ji Soo Choi MD, PhD, Abstract Co-Author: Nothing to Disclose

To compare the diagnostic performance of digital breast tomosynthesis (DBT) with that of conventional full-field digital mammography (FFDM) in a population of screening US-detected breast cancers.

From January 2013 to June 2013, 865 women underwent both conventional FFDM and DBT imaging. Among them, 84 patients were classified as having screening US-detected cancers, which were initially detected by screening US in asymptomatic patients with negatively interpreted mammography. The cases with retrospectively seen, overt mammographic findings or incomplete datasets were excluded (n = 43). In the rest 41 women with US-detected mammographically occult cancers (hereafter, UDMOCs), three radiologists independently described the lesion type and location of the most actionable findings in both FFDM and DBT with 4-week interval, without knowledge of US findings. The lesion type was divided into: negative, mass, asymmetry, focal asymmetry, calcifications only, single view mass and mass with calcifications. The visibility score from 0 to 2 was given to each case.

Diagnostic performance when sum of visibility scores of three radiologists was equal to or more than score 4 was significantly different between DBT and FFDM ( 53.6% vs. 26.8%, P=0.013). Among 41 cases, all three readers detected cancers in 11 cases with DBT and 1 case with FFDM and no reader detected cancers in 6 cases with DBT and 16 cases with FFDM. We found significant difference of diagnostic performance between DBT and FFDM in aspect of the individual visibility in two of three radiologists (63.4% vs. 31.7% for reader 1, p=0.008; 43.9% vs. 9.7% for reader 3, p =0.001). The dominant lesion type was “mass” on DBT (50%) and “focal asymmetry” on FFDM (57.1%).

This study has demonstrated improved diagnostic performance of DBT compared to FFDM in patients with UDMOCs and increased accuracy for mass characterization of DBT compared to FFDM.

When DBT is added to FFDM, cancer detection and mass characterization could be improved in the interpretation of mammographic abnormalities in women with mammographically occult breast cancer.

Nam, K, Han, B, Ko, E, Choi, J, Comparison of Breast Digital Tomosynthesis and Full-field Digital Mammography: Ultrasonography Detected Breast Cancer.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14016735.html