Rubina Noemi Cavallin, Presenter: Nothing to Disclose

Claudio Losio MD, Abstract Co-Author: Nothing to Disclose

Marta Maria Panzeri, Abstract Co-Author: Nothing to Disclose

Elena Venturini MD, Abstract Co-Author: Nothing to Disclose

Giulia Cristel MD, Abstract Co-Author: Nothing to Disclose

Alessandro Del Maschio MD, Abstract Co-Author: Nothing to Disclose

Despite accounting for only 3% of mammographically detected findings, architectural distortion (AD) may be caused by a wide range of benign and malignant breast lesions, and it is the 3rd most common presentation of non palpable breast cancer. Because of its indefinite mammographic and ultrasonographic features, percutaneous or surgical biopsy is mandatory. In our study we evaluated the potential role of dynamic Breast MRI including diffusion-weighted imaging (DWI) as a problem-solving tool in mammographically and ultrasonographically detected AD

Out of 232 patients undergoing MRI for problem solving, 34 were examined for a mammographic or ultrasonographic AD. MRI (1.5T) included T2-TSE sequences, dynamic study and DWI (b-values: 0, 900 s/mm2). For each lesion detected we evaluated morphology, dynamic and diffusion patterns and final histopathological result. A cut-off ADC value differentiating benign from malignant breast lesions had been previously established in a large population of women. The difference between the mean Apparent Diffusion Coefficient (ADC) values and the mean T2 signal in malignant and benign findings was evaluated with Mann-Whitney U test. Univariate and multivariate analyses of ADC values, T2 signal and time-enhancement curves (T-Ec)  were performed for prediction of malignancy.  

MRI confirmed all 38 known findings. At histopathology 25 lesions were benign and 13 malignant. The most represented T-Ec in malignant AD were type 3 (n=7) and 2 (n=3), while no benign lesions showed a washout kinetic. Univariate and multivariate analysis showed that T-Ec were significant predictors of malignancy (p<0.05). Quantitative DWI assessed a mean ADC value of 1.31x10-3mm2/s in benign lesions, higher than in malignant lesions (1.20x10-3mm2/s, p<0.05). ADC values were significant for predicting malignancy with use of univariate (p<0.05) but not multivariate analysis (p=0.08). T2 values were not predictors of malignancy.  

Time-enhancement curves were the most predictive MRI feature to distinguish benign from malignant AD. The contribution of DWI to their differential diagnosis is limited due to ADC borderline values. MRI low negative predictive value, however, suggests to avoid MRI to strengthen the diagnosis after a benign core biopsy.  

MRI could not replace breast biopsy to confirm the nature of architectural distortions.  

Cavallin, R, Losio, C, Panzeri, M, Venturini, E, Cristel, G, Del Maschio, A, Breast MRI as a Problem-solving Tool in the Evaluation of Mammographically and Ultrasonographically Detected Architectural Distortions: Are There Any Predictive Parameters?.  Radiological Society of North America 2013 Scientific Assembly and Annual Meeting, December 1 - December 6, 2013 ,Chicago IL. http://archive.rsna.org/2013/13027477.html