Güner Gülbanu, Turgan Ahmet Harun, Emir Sevde Nur
Health Sciences University, Umraniye Training and Research Hospital, Department of Radiology, Istanbul, Turkey.
Health Sciences University, Umraniye Training and Research Hospital, Department of Radiology, Istanbul, Turkey.
Eur J Radiol. 2025 Sep;190:112211. doi: 10.1016/j.ejrad.2025.112211. Epub 2025 Jun 1.
To evaluate the diagnostic efficacy of abbreviated breast magnetic resonance imaging (AB-MRI) in aiding the differentiation of intraductal papilloma (IDP) from ductal secretion, as compared to full diagnostic breast magnetic resonance imaging (FDB-MRI).
This retrospective study included patients who underwent breast MRI between June 2016 and January 2024 due to inconclusive ultrasonographic findings suggestive of IDP or ductal secretions, A total of 62 patients were analyzed with histopathological results used as the reference standard. Two radiologists independently reviewed the AB-MRI protocols in a blinded manner. These protocols included fat-suppressed precontrast T1, first postcontrast T1, and subtracted images derived from the FDB-MRI. After a one-week interval, the same radiologists evaluated the FDB-MRI examinations in a separate session, blinded to the AB-MRI results. Diagnostic performance metrics including sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated. Interobserver agreement for AB-MRI, as well as agreement between AB-MRI and FDB-MRI interpretations, were analyzed using Cohen's kappa test.
The mean age of the 62 patients was 46.05 ± 11.59 years, Histopathological results identified 48 IDPs, 6 cases of adenosis and fibrosis, and 8 cases of ductal secretions. The FDB-MRI demonstrated high sensitivity (87.5 %) and PPV (85.7 %) for IDP detection but limited specificity (50 %) and NPV (53.85 %). The comparison between FDB-MRI and pathology did not reveal a statistically significant difference (p > 0.05). The diagnostic performance of AB-MRI was observer-dependent; for Observer 1, sensitivity and specificity were 75.0 % and 57.1 %, respectively, with no significant difference compared to pathology (p = 0.0525) or FDB-MRI (p = 0.124). For Observer 2, sensitivity and specificity were 66.67 % and 57.14 %, also not significantly different from pathology (p = 0.1945) or FDB-MRI (p = 0.347). Interobserver agreement was high (Cohen's Kappa = 0.79) CONCLUSION: The AB-MRI protocol demonstrated moderate-to-high diagnostic performance and high interobserver agreement as a time-efficient alternative. While the FDB-MRI remains superior in sensitivity and diagnostic accuracy, AB-MRI offers a reasonable alternative in clinical scenarios where shorter imaging time is prioritized, such as in the differentiation of IDPs from ductal secretions.
