According to Dr. Monica Morrow (the Tom Brady of breast oncology):
MRI finds two to three times more disease in observed rates of local recurrence in patients selected (for breast conserving surgery)without MRI. This results in increased mastectomy rate for questionable patient benefit. To date neither short term surgical outcomes nor long term local control or contralateral breast cancer rates are impoved with MRI.
Essentially, MRI does not improve surgical outcomes, leads to a higher rate of unnecessary mastectomy, and is extremely expensive (about $1600 a pop, out of pocket). Seems like a slam dunk---time to start getting away from routine pre-operative MRI's for breast cancer.
I had a patient recently who had had a mammogram which showed multiple concerning areas of pleomorphic calcifications diffusely throughout the breast. I sent her for stereotactic needle biopsy of all the spots but the radiologist called and said he wanted to get an MRI first. I was in the middle of something, so I hastily agreed, assuming he had a good enough reason. Well the MRI showed... multiple suspicious areas of pleomorphic calcifications. All the subsequent core biopsies were positive for DCIS. She ended up getting a mastectomy. In retrospect, the MRI added absolutely nothing to the case. We knew she had multicentric/multifocal disease based on mammography alone. The MRI simply (expensively) confirmed what we already knew. It was wasteful, delayed defintive surgery, and just added more stress and cost to an already charged situation.
There are situations where MRI could potentially be useful (to help resolve discordances betwen conventional imaging and physical exam, to assess response to neoadjuvant chemotherapy, inflammatory breast cancer, patients with positive axillary lymph nodes and occult primary breast cancers, some patients with hereditary breast cancer syndromes) but at this point in time there is not nearly enough evidence to support its routine use in most patients with breast cancer...