RESUMO
OBJECTIVE: The purpose of this study is to provide a more accurate estimation of the radiation dose of contrast-enhanced spectral mammography (CESM) relative to that of 2D digital mammography and tomosynthesis using phantom and patient data and an accepted dosimetry protocol that eliminates vendor-specific average glandular dose (AGD) estimates while including breast density. MATERIALS AND METHODS: Patient and phantom AGD estimation was performed using two vendors (system 1 and system 2) in five imaging modes, including 2D, 3D, and CESM imaging. Patient AGD was retrospectively estimated from 45 patients who underwent mammography with all imaging modes during 2012-2016. Patient and phantom AGD were estimated using accepted European and International Atomic Energy Agency protocols for dosimetry and were compared across imaging modes using a paired t test with Bonferroni correction. RESULTS: Phantom data showed that the imaging modes with the lowest to highest AGDs were system 1 2D, followed by system 2 2D and system 2 3D, which had comparable values (p = 0.6), followed by system 1 CESM, and then by system 2 2D plus 3D. One hundred eighty views in 45 patients showed that the system 1 CESM AGD was 1.8 times greater than the system 1 2D AGD (p < 0.001), 1.2 times greater than the system 2 2D AGD (p < 0.001), 1.2 times greater than the system 2 3D AGD (p < 0.001), and 0.6 times less than the system 2 2D plus 3D AGD (p < 0.001). CONCLUSION: The CESM dose for system 1 is within an acceptable range as compared with other commonly performed mammographic examinations and should not preclude its use as a diagnostic breast imaging tool.
