We retrospectively analyzed 36 cases of FFDM or DBT detected primary AD enrolled for CEDM between August 2021 and September 2022 (total of 117 CEDM examinations).

Lesions that fulfilled following parameters were included in our analysis: 1) AD in patients with BI-RADS category 4 or 5; 2) accessible pathologic data from either image-guided or surgical biopsies; and 3) at least one year of imaging or clinical follow-up. Mammographic masses with spiculation (BI-RADS 5) and lesions with AD linked to established benign causes, such as a stable postsurgical scar, were excluded (BI-RADS 2 lesions) in order to solely include the lesions with primary AD.

ADs after CEDM were also categorized according to the BI-RADS lexicon; ADs categorized as BI-RADS 4 and 5 were referred for biopsy or surgical excision (Figure 1.). Histopathologic results or follow-up of at least one year were used as the reference standard. Statistical analysis was performed using MedCalc® Statistical Software version 20.111 (MedCalc Software Ltd, Ostend, Belgium; https://www.medcalc.org; 2022).

All participants gave their informed consent prior beeing included in the study (Ethical approval number: 03-2673-2/22).

CEDM imaging protocol: An iodine based contrast agent is intravenously administered two minutes prior to image acquisition. Afterwards, a dual-energy mammography examination is conducted in the craniocaudal (CC) and mediolateral oblique (MLO) views of both breasts. Additional delayed images, after 8 minutes, were made in CC and MLO views of the breast with AD. Finally, for radiological analysis were available low-energy (LE) images, which are comparable to a traditional FFDM images, and recombined (iodine) images (in which areas of contrast uptake can be appreciated).

Readings were conducted using dedicated workstations that were calibrated for regulated, ambient lighting settings, and, if avaible, compared with earlier examinations, and the standard screening or diagnostic mammography that revealed the initial AD.