Background

• State-of-the-art mammography, including digital mammography (DM), is of pivotal importance to achieve early diagnosis of breast cancer, and the only imaging tool proved to reduce mortality [1]. However, the overlap of fibro-glandular tissue in dense breasts may affect sensitivity and specificity of DM by masking or mimicking a cancer [2].
• Digital Breast Tomosynthesis (DBT) is a novel imaging tool aimed to overcome tissue overlap, since it is based on a moving x-ray source and a digital detector used to obtain a three-dimensional (3D) volume of thin section data. Images are then reconstructed with proper algorithms to obtain a set of thin image sections parallel to the breast platform [3].
• DBT is still under investigation. This technique is expected to help radiologists in improving: (i) lesions detectability; (ii) the characterization of masses and architectural distortions (especially in dense breasts); (iii) the definition of lesions in terms of dimensions and margins [2-3]. 

Purpose

The aim of our study was to compare DBT and DM in terms of: (i) diagnostic yield (DY) for cancer (primary endpoint); (ii) the agreement in assessing lesion size (secondary endpoint).