Synthesized Mammography. The New Standard of Care When Screening for Breast Cancer with Digital Breast Tomosynthesis?

Emily B. Ambinder, Susan Harvey, Babita Panigrahi, Ximin Li, Ryan Woods

Research output: Contribution to journalArticlepeer-review

Abstract

Rationale and Objectives: This study aims to evaluate the screening performance of digital breast tomosynthesis (DBT) combined with synthesized mammography (SM) vs combined with full-field digital mammography (FFDM). Materials and Methods: We retrospectively reviewed all screening studies utilizing FFDM + DBT (n = 7845) and SM + DBT (n = 14,776) between April 1, 2013, and February 15, 2016. Recall rate, biopsy rate, positive predictive value 1 (PPV1), positive predictive value 3 (PPV3), and cancer detection rate (CDR) were compared between the two groups. A generalized linear mixed model specifying the reading radiologist as the random effect and controlling for age was used to compare clinical outcomes between the two groups. Results: The overall recall rate was significantly lower in the SM + DBT cohort compared to the FFDM + DBT cohort (7.06% vs 7.63%, . P = .04). There was no difference in biopsy rate, PPV1, PPV3, or CDR between the two groups. Conclusions: When DBT is performed for screening, the use of SM rather than acquiring an additional FFDM has no significant effect on biopsy rate, PPV1, PPV3, or CDR. We found a decrease in recall rate in the SM + DBT group, which may be related to the learning curve of interpreting DBT. These findings support the use of SM for patients undergoing screening with DBT.

Original languageEnglish (US)
JournalAcademic Radiology
DOIs
StateAccepted/In press - Jan 1 2018

Keywords

  • Breast neoplasms
  • Early detection of cancer
  • Mammography

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging

Fingerprint Dive into the research topics of 'Synthesized Mammography. The New Standard of Care When Screening for Breast Cancer with Digital Breast Tomosynthesis?'. Together they form a unique fingerprint.

Cite this