Novel scoring system using contrast-enhanced mammography to differentiate benign and malignant architectural distortion

Abstract

Abstract Background Incidence of malignancy in architectural distortion (AD) is variable, and as a standard of care, all the architectural distortions are biopsied. It is important to identify benign and malignant characteristics of ADs on imaging to predict the probability of malignancies as well as to establish radiological–pathological concordance. The aim of our study was to analyse the utility of contrast-enhanced mammography (CEM) in primary AD, to describe and evaluate the accuracy of CEM parameters and derive a novel scoring system to differentiate benign and malignant AD. Results Forty-six patients with 49 ADs were evaluated with a median age, 51 years (age range—34–72 years) and all were women. Of the 49 ADs, 65.3% (32/49) were benign, and 34.69% (17/49) were malignant. The CEM parameters which predicted malignancy were moderate or marked density, size > 0.95 cm, CT ratio > 1.31, mass morphology and washout kinetics. The absence of enhancement and enhancing AD with AD score of 0 had a 100% NPV for malignancy. Enhancing architectural distortion with a score of 3 or more had 100% PPV for malignancy. A score of 1 or 2 was indeterminate, and an individualised decision was recommended. Conclusions In primary AD, several CEM features have significantly predicted malignancy when the score is 3 and above. A biopsy may be avoided in non-enhancing AD and enhancing AD with a score of 0 due to high NPV.