Francesca Fornasa MD, Presenter: Nothing to Disclose

Lucia Pinali, Abstract Co-Author: Nothing to Disclose

Anna Gasparini, Abstract Co-Author: Nothing to Disclose

Ermanno Toniolli, Abstract Co-Author: Nothing to Disclose

Angelo Dibenedetto, Abstract Co-Author: Nothing to Disclose

Stefania Montemezzi, Abstract Co-Author: Nothing to Disclose

To assess the accuracy of diffusion-weighted MR imaging (DWI) in distinguishing between benign and malignant focal breast lesions.

93 female patients, each with one focal breast lesion at least 7 mm in diameter, underwent both dynamic MRI (basal T1 fast spin-echo and T2 fat-suppressed sequences; 6 gradient-echo T1 sequences every minute after intravenous injection of Gadolinium) and DWI (one parallel-imaging echo-planar sequence with b values of 0 and 800 sec/mm2), with measurement of the apparent diffusion coefficient (ADC). A pathologic diagnosis (by means of fine-needle aspiration cytology, core biopsy and/or histologic examination of a surgical specimen) was eventually obtained in all cases. The diagnostic performance of DWI (adopting the cutoff values of 1.48 and 1.52 x 10-3 mm2/sec) was evaluated using pathology as a gold standard.  

At pathology, 49 lesions proved benign and 44 malignant. DWI performed only slightly (overall accuracy: 91.4 versus 92.5%; difference not statistically significant) worse than MRI in differentiating benign and malignant lesions; depending on the cutoff ADC value chosen, its sensitivity ranged from 88.6 to 100% and its specificity from 83.7 to 93.9%. With both cutoff values, the difference between ADC values measured in benign and malignant lesions was highly statistically significant (p < 0.001) at the Mann-Whitney test. A false-positive DWI result, in which a very low (1.32 x 10-3 mm2/sec) ADC value was measured, occurred in a case of steatonecrosis. We hypothesize that this finding may be caused to the difficulty of water movement in a highly hydrophobic extracellular environment due to the presence of lipidic molecules.  

Although it cannot yet avoid pathologic characterization, DWI is a well-performing tool in differentiating benign and malignant focal breast lesions.

This study, using pathology as a gold standard, demonstrates the usefulness of DWI in characterizing focal breast lesions. A hypothesis is advanced for a false-positive result in steatonecrosis.

Fornasa, F, Pinali, L, Gasparini, A, Toniolli, E, Dibenedetto, A, Montemezzi, S, Diffusion-weighted Magnetic Resonance Imaging Characterization of Focal Breast Lesions: Analysis of 93 Cases with Pathologic Diagnosis.  Radiological Society of North America 2009 Scientific Assembly and Annual Meeting, November 29 - December 4, 2009 ,Chicago IL. http://archive.rsna.org/2009/8003841.html