US and MRI in the evaluation of mammographic BI-RADS 4 and 5 microcalcifications

Abstract

Purpose: The aim of this study was to assess diagnostic accuracies of ultrasonography (US) and magnetic resonance imaging (MRI) in lesions that manifest as mammographic BI-RADS 4 and 5 microcalcifications, in the setting of conjoined use of mammography, US, and MRI.

Methods: Patients with mammographic BI-RADS 4 or 5 microcalcifications, without additional findings, were included in this prospective study. All patients subsequently underwent breast US and MRI. Histopathologic diagnosis, obtained by US-guided core-needle biopsy or surgical excision, served as a reference standard. Diagnostic accuracies of US and MRI were calculated, and positive predictive value for different MRI BI-RADS imaging features were determined.

Results: The study group consisted of 113 women with 125 areas of suspicious microcalcifications. MRI reached sensitivity, specificity, positive predictive value 3 (PPV3), and negative predictive value (NPV) of 100%, 70.1%, 67.6%, and 100%, respectively. Statistically significant differences in MRI morphologic features and kinetic enhancement curves were observed between malignant and benign microcalcifications. Sensitivity, specificity, PPV3, and NPV for US were: 85.4%, 66.2%, 61.2%, and 87.9%. There was statistically significant difference in presentation of malignant and benign microcalcifications at US.

Conclusion: In the setting of conjoined use of mammography, US, and MRI, MRI can reliably exclude malignancy in suspicious microcalcifications. Thus, negative MRI findings may influence the decision to biopsy the microcalcifications.

Figure 1 a, b

Mammography, mediolateral oblique (a) and craniocaudal (b) projections: fine pleomorphic and linear branching microcalcifications, in segmental distribution, in the upper outer quadrant of the right breast. Multiple lobulated densities are seen in all quadrants.

Figure 2

Ultrasound: microcalcifications were in B-mode visible as hyperechoic dots (yellow arrows) surrounded by heterogeneous irregular area. Multiple large cysts in all quadrants (asterisk) correlated with lobulated densities visible on mammography. Circles with “x” and “+” marks remained from elastography measurements, which were not analyzed in this study.

Figure 3 a, b

Subtracted postcontrast MRI (a) shows non-mass lesion of the right breast, with clumped internal enhancement and regional distribution. Dynamic-kinetic curve with ROI inside non-mass lesion had rapid enhancement in initial phase with plateau in the delayed phase (b). Second-look US-guided core needle biopsy revealed atypical ductal hyperplasia. Suspicious imaging findings indicated open surgical excision. Upon mastectomy microinvasive ductal carcinoma was proven.