Breast MRI of pure ductal carcinoma in situ: sensitivity of diagnosis and influence of lesion characteristics

Objectives: The purpose of the study was to evaluate the sensitivity of breast MRI in the detection of pure DCIS and to analyze the influence of lesion type and nuclear grade.

Methods: 58 consecutive patients with pathologically proven pure DCIS and preoperatively performed breast MRI were retrospectively reviewed and analyzed. Sensitivities in the detection of DCIS were calculated for MRI and mammography (Mx). Influence of MRI lesion type and nuclear grading on DCIS diagnosis was evaluated.

Results: MRI detected pure DCIS with a sensitivity of 79.3%. The sensitivity of Mx was lower (69%), but the difference was not statistically significant (p=0.345). 46.2% of the DCIS presented as enhancing mass and 53.8% as non-mass-like enhancement (NMLE). None of the masses but 21.4% (n=6) of the NMLE were underestimated as probably benign (BI-RADS 3). MRI measured lesion sizes showed a moderate correlation (r=0.74) with histopathologically measured lesion sizes. MRI detection rate of DCIS decreased significantly (p=0.0458) with increasing nuclear grade. Calculated sensitivities were 100% for low-grade DCIS, 84.6% for intermediate-grade DCIS, and 66.7% for high-grade DCIS.

Conclusions: In this study MRI could detect pure DCIS more sensitively than Mx. Despite of missing statistically significance preoperative MRI seems to be helpful in patients with DCIS who are eligible for breast conservation. This applies in particular to patients with non-high-grade DCIS because those were significantly more often positive on MRI and significantly more often negative on Mx. Misinterpretation occurs especially in cases of NMLE and high-grade DCIS and therefore a correlation with Mx is also recommended.