Abstract Archives of the RSNA, 2012
LL-BRS-WE4D
Comparability of Digital Breast Tomosynthesis with Conventional Mammography, Ultrasonography and MRI for Tumor Measurement
Scientific Informal (Poster) Presentations
Presented on November 28, 2012
Presented as part of LL-BRS-WEPM: Breast Imaging Afternoon CME Posters
Melissa Angeline Durand MD, Abstract Co-Author: Nothing to Disclose
Christine Chen MD, Presenter: Nothing to Disclose
Reni Simov Butler MD, Abstract Co-Author: Nothing to Disclose
Liane Elizabeth Philpotts MD, Abstract Co-Author: Nothing to Disclose
To assess the accuracy of tumor measurement utilizing digital breast tomosynthesis (DBT), define the optimal parameters for measurement with DBT, and evaluate DBT's accuracy in comparison with conventional 2D mammography (2D), ultrasonography and MRI, in patients with breast cancer.
Over a period of 6 months (August 2011 through February 2012), 26 breast cancer patients for whom DBT was performed completed surgical treatment and were evaluated for this study. Patients were excluded if DBT was performed after a biopsy was performed.
Two radiologists with 13 years of experience (RB) and 18 years of experience (LP) were blinded to final pathology measurements and independently assessed 19 patients with DBT and 2D. Tumor measurements in three dimensions in CC and MLO projections were recorded for DBT and 2D. On DBT, measurements of tumors without spicules and including spicules were recorded. The largest tumor measurement without and with spicules in either projection on DBT and 2D, was then compared to the largest measurement on ultrasound, MRI and final pathology.
Pearson's correlation coefficients (R) were calculated to compare the tumor measurements across the various modalities.
DBT tumor measurements including spicules matched final pathology (R 0.58) more closely than measurements without spicules (R 0.45). Both methods of DBT measurement were better matched to pathology measurements compared to 2D (R 0.58 and 0.45 vs. R 0.42). Across all modalities, ultrasound measurements most closely matched pathology measurements (R 0.74).
DBT allows for more accurate tumor measurement compared to 2D. In addition, DBT measurement including tumor spicules more closely reflects final pathology measurements than measurement without tumor spicules. This may be due to less tissue superimposition and increased lesion conspicuity on DBT compared to 2D.
Improved accuracy of pre-surgical tumor measurements using digital breast tomosynthesis may aid optimization of surgical and treatment planning.
Durand, M,
Chen, C,
Butler, R,
Philpotts, L,
Comparability of Digital Breast Tomosynthesis with Conventional Mammography, Ultrasonography and MRI for Tumor Measurement. Radiological Society of North America 2012 Scientific Assembly and Annual Meeting, November 25 - November 30, 2012 ,Chicago IL.
http://archive.rsna.org/2012/12043738.html