Comparative Accuracy of Preoperative Tumor Size Assessment on Ultrasonography and Magnetic Resonance Imaging in Ductal Carcinoma In Situ

Purpose: The purpose of this study was to compare the accuracy of preoperative breast tumor size measured by ultrasonography (US) and magnetic resonance imaging (MRI) in patients with ductal carcinoma in situ (DCIS). Methods: Medical records of 74 patients postoperatively diagnosed with DCIS were retrospectively analyzed. Tumor size measurements obtained using the two imaging modalities were compared for accuracy with those obtained during the final pathologic examination. Patients with only microcalcification on imaging were excluded. Results: For all patients, Lin’s concordance correlation coefficient (CCC) of MRI was 0.725, which was more accurate than 0.670 of US. In subgroup analysis, CCC of US was 0.757, more accurate than 0.697 of MRI in premenopausal DCIS patients. Background parenchymal enhancement (BPE) was the only risk factor deteriorating the accuracy of US and MRI examinations. Moderate and marked BPE was associated with the inaccurate tumor size estimations in both US and MRI (odds ratio [OR]:2.23, 95% confidence interval [CI]=1.12−3.52, p=0.001 in US, OR:8.16, 95% CI=1.17−15.2, p=0.031 in MRI). Conclusion: MRI was more accurate for measuring tumor size in patients with DCIS. Moderate and marked BPE was a risk factor that prevented accurate measurement of preoperative tumor size. In premenopausal patients, US would help measure tumor size accurately.