BI-RADS-MRI terminology and evaluation of intraductal carcinoma and ductal carcinoma in situ.

There has been dramatic progress in MRI technology during the past 20 years, and the rate of detection and diagnostic accuracy in regard to intraductal carcinoma and ductal carcinoma in situ (DCIS) have been increasing. First, we present MRI images of intraductal carcinomas and the terminology in the second edition of the BI-RADS-MRI to describe them. Next, we examined the data in our institution in regard to the following: (1) the DCIS detection rate, (2) the proportions of breast cancer and DCIS in MR-guided vacuum-assisted biopsies (VAB), (3) evaluation of the extent of intraductal carcinoma, and (4) diagnosis of extension of intraductal carcinoma into the nipple. MR images were acquired by performing a 1-min interval dynamic study with a 1.5-T MR scanner. The same radiologist evaluated the MRI in all of the cases. MR-guided VAB was performed by using a commercially available biopsy system. (1) The DCIS detection rate was 95% (148/156), and the DCIS lesions that MRI was unable to detect were low grade in 5 cases, intermediate grade in 2 cases, and high grade in 1 case. (2) The proportion of MR-guided VAB specimens that were breast cancer was 36% (71/200), and the proportion of breast cancers that were DCIS was 82% (58/71). (3) The proportion of margin-positive specimens in the 100 breasts in which breast-conserving surgery was performed was 11% (11/100), and the proportion in which MRI was the cause of the margin being positive was a mere 4% (4/100). (4) The positive predictive value of periductal enhancement and linear enhancement for a diagnosis of intraductal extension into the nipple was 83% (10/12) and 43% (3/7), respectively. Their negative predictive value was 100% (58/58). We concluded that MRI is a very useful examination for the diagnosis of intraductal carcinoma, and that it is important to become thoroughly familiar with the BI-RADS-MRI terminology and accurate interpretation methods.