Synthetic 2D Mammography Versus Standard 2D Digital Mammography: A Diagnostic Test Accuracy Systematic Review and Meta-Analysis.

Background: The use of synthetic 2-dimensional mammography (SM) with 3D digital breast tomosynthesis (DBT) in place of standard 2D digital mammography (DM) may reduce radiation dose without sacrificing accuracy. Objective: To compare the diagnostic accuracy of SM, DM, SM with DBT, and DM with DBT for breast cancer detection. Evidence Acquisition: A search of MEDLINE, EMBASE, and the Cochrane database was performed for relevant articles published up to September 2019. Studies included compared the diagnostic accuracy of SM versus DM and SM+DBT versus DM+DBT for breast cancer detection. Relevant study data metrics and risk of bias were assessed. A bivariate random-effects meta-analysis and meta-regression were performed to assess diagnostic accuracy (PROSPERO CRD42020150737). Evidence Synthesis: Thirteen studies reporting on 201,304 patients (7,252 with breast cancer) were included. The SM group (six studies, 20,728 patients, 724 with breast cancer) had 76% sensitivity (95%-confidence interval [CI] 66%-83%) and 91% specificity (CI 81%-96%). The DM group (nine studies, 52,082 patients, 2,249 with breast cancer) had 74% sensitivity (CI 65%-81%) and 86% specificity (CI 74%-93%). The SM+DBT group (nine studies, 64,224 patients, 2,149 with breast cancer) had 83% sensitivity (CI 78%-87%) and 92% specificity (CI 84%-96%). The DM+DBT group (nine studies, 64,270 patients, 2,130 with breast cancer), had 84% sensitivity (CI 79%-87%) and 90% specificity (CI 81%-95%). Meta-regression models did not reveal significant differences in accuracy between the SM and DM groups (p=0.247-0.768). Conclusions: SM and SM+DBT demonstrated a comparable diagnostic accuracy to DM and DM+DBT, respectively. Clinical Impact: The findings support the implementation of SM in place of standard DM for mammographic imaging of the breasts. This may lead to an overall reduction in radiation exposure.