Francesca Proulx MD, Presenter: Nothing to Disclose

Benoit Delphin Mesurolle MD, Abstract Co-Author: Nothing to Disclose

Attila Omeroglu, Abstract Co-Author: Nothing to Disclose

Brenda Macgibbon PhD, Abstract Co-Author: Nothing to Disclose

Jose A. Correa PhD, Abstract Co-Author: Nothing to Disclose

Sepideh Babaei MD, Abstract Co-Author: Nothing to Disclose

Sarkis Meterissian, Abstract Co-Author: Nothing to Disclose

The purpose of this study is (1) to evaluate the accuracy of pre-operative breast MRI and mammography in determining the tumour extent (size) of DCIS compared to the histopathological results, and (2) to determine whether the breast density, nuclear grade and comedo-necrosis can affect the ability of mammography and MRI to accurately assess the size of DCIS.

We performed a retrospective analysis of proven cases of pure DCIS of the breast diagnosed between December 2007 and January 2011 in 79 patients (age range: 37-77 years, mean age: 56.5 years). Mammographic and MRI features were retrospectively analyzed by 2 breast radiologists following the BI-RADS lexicon and the size of the lesion was recorded. The histopathologic DCIS size, type and grade as well as the presence of comedonecrosis were also reviewed. The MRI and mammographic size of DCIS were compared with the histopathologic size (Pearson’s correlation) and the influence of the other variables was evaluated with the chi-squared test and parametric t-test.

Of the 79 DCIS cases, 72 were detected by mammography and 67 by MRI. The overall mean histopathologic size of DCIS was 2.56 cm. The Pearson’s coefficient showed a better correlation of the size between MRI and histopathology (0.443) than mammography and histopathology (0.261) (p=0.027). Mammography underestimated the size by at least 1 cm in 29 % of the cases, was accurate in 46 % and overestimated the size by at least 1 cm in 25 %. MRI overestimated the size by at least 1 cm in 28% of the cases, was accurate in 51% and underestimated the size by at least 1 cm in 21 %. Neither comedo-necrosis nor breast density had a significant effect on either the accuracies of the mammography or MRI measurements. However, nuclear grade had an effect on the MRI size measurements when compared to the histopathology size as measured by the chi-squared test (p=0.001).

MRI is better than mammography at assessing the size of DCIS when the lesion is seen.

Contrast-enhanced MRI is a tool to help capture the size and extent of DCIS when identified and can be indicated in the pre-operative setting.

Proulx, F, Mesurolle, B, Omeroglu, A, Macgibbon, B, Correa, J, Babaei, S, Meterissian, S, Value of Preoperative Breast MRI for the Size Assessment of Ductal Carcinoma In Situ (DCIS).  Radiological Society of North America 2012 Scientific Assembly and Annual Meeting, November 25 - November 30, 2012 ,Chicago IL. http://archive.rsna.org/2012/12022557.html