This article is more than 5 years old. Information may no longer be current.
Preoperative breast MRI: Too much of a good thing?
Click Here to Manage Email Alerts
The use of preoperative breast MRI has increased in the United States and Canada.
In the absence of data demonstrating that its use improves survival of women diagnosed with invasive breast cancer, it is relevant — perhaps imperative — to examine other downstream consequences, including those that are undesirable.
In this population-based retrospective cohort study, Arnaout and colleagues examined breast MRI use from 2003 to 2012 in the province of Ontario. They found 14.8% of patients underwent the test.
MRI use increased eightfold, from 3% to 24% of all women, annually over the decade. Factors associated with MRI use included younger age, higher socioeconomic status, higher Charlson comorbidity score, surgery performed in a teaching hospital and fewer years of surgeon experience.
The use of preoperative breast MRI was associated with a higher likelihood of breast imaging, breast biopsies and imaging to assess for distant metastatic disease after diagnosis, as well as contralateral prophylactic mastectomy and longer waiting time to definitive breast surgery.
Again, there has not been any demonstration that preoperative breast MRI saves lives. Seemingly, we are at the intersection of “seek and ye shall find” and “primum non nocere” — first, do no harm.
One may hypothesize why younger women of higher socioeconomic status in particular were more likely to undergo preoperative breast MRI — fear of missing multifocal disease in denser breast tissue, and the potentially greater use of the test owing to less concern about cost. My own observation is that preoperative breast MRI often sets a snowball effect in motion characterized by a series of events that can culminate in unwarranted contralateral mastectomy.
It can go something like this:
1) An understandably vulnerable patient with a T1c invasive ductal carcinoma has MRI revealing indeterminate density(ies) distant from the already diagnosed invasive carcinoma in either the ipsilateral or contralateral breast — or both — for which additional imaging and/or biopsy is recommended.
2) Stereotactic core needle biopsy(ies) of indeterminate or suspicious lesion(s) is(are) performed.
3) The patient waits anxiously for the pathology report.
4) The patient considers what possible course she might take if the pathologic findings warrant unilateral or bilateral mastectomy.
5) Well-intentioned friends and relatives offer reassuring counsel — perhaps unsolicited — about the favorable cosmetic results of bilateral breast reconstruction for this patient who may ultimately be well served by unilateral breast conserving surgery, an idea that may be reinforced by plastic surgical consultation. This may be further confounded by the patient in the waiting room who supportively confides her total lack of regret for having had bilateral mastectomy with prophylactic contralateral surgery, not mentioning her own deleterious BRCA1 mutation.
6) By the time the pathology report returns with benign results, the patient — having already travelled down a “slippery slope of angst” — has convinced herself “not to take any chances” by doing anything less than bilateral mastectomy.
My esteemed colleague, Monica Morrow, MD, FACS, commented 12 years ago: “Whether MRI will result in a meaningful clinical benefit is an open question, and one that should have been answered prior to its widespread adoption.”
She continued: “Evidence of clinical benefit in terms of OS, DFS or quality of life is the established standard for new therapeutic modalities in breast cancer. When imaging studies are used to select therapy, the same rigorous standards must be applied.”
For the patient described above, an arguably unnecessary preoperative breast MRI sent a snowball rolling down a hill and, indeed, unleashed an avalanche.
Whether patients or surgeons are driving increased use of routine preoperative breast MRI, current data argue for a much more conservative approach. With an apparent scenario of technology penetration getting ahead of outcomes data, it behooves us to remember: “primum non nocere.”
References:
Morrow M. JAMA. 2004;doi:10.1001/jama.292.22.2779.
Pilewskie M and King TA. Cancer. 2014;doi:10.1002/cncr.28700.
Wang SY, et al. J Clin Oncol. 2016;doi:10.1200/JCO.2015.62.9741.
For more information:
Andrew D. Seidman, MD, is attending physician on the breast medicine service at Memorial Sloan Kettering Cancer Center and professor of medicine at Weill Cornell Medicine. He also serves as a HemOnc Today Editorial Board member. He can be reached at seidmana@mskcc.org.
Disclosure: Seidman reports no relevant financial disclosures.