Background: Our aim was to compare the accuracy of magnetic resonance imaging (MRI) and ultrasonography (US) in measuring the size of invasive breast cancer (IBC) and carcinoma in situ (CIS). We also examined the utility of routinely performing MRI in addition to US before breast-conserving surgery (BCS).
Patients and methods: Data from 1558 consecutive patients diagnosed with IBC and/or CIS between 2003 and 2005 were reviewed. For comparing the accuracy of US and MRI, paired t test was done comparing pathologic and imaging (US and MRI) tumor size in 821 patients who received both breast US and MRI. In instance of attempted BCS (n = 794), operative approach, resection margins, and clinical outcomes of non-MRI and MRI groups were compared.
Results: For CIS, IBC without CIS, and IBC with CIS, MRI was more accurate in estimating tumor size than US. When BCS was attempted (n = 794), the rate of tumor involvement in initial resection margins did not differ between non-MRI and MRI groups (23.0% and 23.4%, P = .926). Similarly, rates of re-excision (13.1% vs 17.5%, P = .130) and conversion to mastectomy (2.3% vs 2.1%, P = .893) were comparable, as were ipsilateral breast tumor recurrence, locoregional recurrence, and disease-free survival (log rank P = .284, .950, and .955, respectively).
Conclusions: Breast MRI provided more accurate estimates of tumor size, correlating better with pathologic tumor size than US for both IBC and CIS. However, no clear benefit in terms of lower re-excision rate, higher breast conservation success, or reduced recurrence emerged for routine use of breast MRI before BCS.