Differentiation of benign and malignant sub-1 cm breast lesions using dynamic contrast enhanced MRI

• Published in: Breast (Edinburgh) Vol. 13; no. 2; pp. 115 - 121
• Main Authors: Gibbs, P, Liney, G.P, Lowry, M, Kneeshaw, P.J, Turnbull, L.W
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier Ltd 01.04.2004
• Subjects: Adult; Aged; Breast cancer; Breast Neoplasms - diagnosis; Breast Neoplasms - pathology; Carcinoma, Ductal - diagnosis; Carcinoma, Ductal - pathology; Contrast enhanced MRI; Female; Humans; Image Enhancement; Logistic Models; Magnetic Resonance Imaging; Mammography; Middle Aged; Retrospective Studies; ROC Curve; Sensitivity and Specificity; Small lesions; Ultrasonography, Mammary; Breast cancer; Small lesions; Contrast enhanced MRI
• ISSN: 0960-9776; 1532-3080
• DOI: 10.1016/j.breast.2003.10.002

The purpose of this work is to assess the additional benefit of MRI-based morphology and quantification of contrast enhancement in the differential diagnosis of sub-1 cm breast lesions. Forty-three women with suspected breast cancer were examined using X-ray mammography, ultrasound mammography, and MRI. Dynamic contrast imaging was performed and relative enhancement at various time-points was calculated. The dynamic data was also processed using a two-compartment pharmacokinetic model. Radiological interpretation of high-resolution post-contrast images revealed a similar accuracy (69%) compared to X-ray mammography (69%) and ultrasound mammography (67%). The best individual parameter calculated from the dynamic images was found to be the exchange rate constant which revealed a diagnostic accuracy of 0.74±0.08. When information from the post-contrast images and dynamic data was combined in a logistic regression model a diagnostic accuracy of 0.92±0.03 was achieved. In conclusion, MR imaging of small breast lesions is feasible and the incorporation of quantitative MR derived parameters is beneficial.