MRI and mammography surveillance of women at increased risk for breast cancer: recommendations using an evidence-based approach

Acad Radiol. 2008 Dec;15(12):1590-5. doi: 10.1016/j.acra.2008.06.006.


Rationale and objectives: To evaluate breast cancer screening with mammography and magnetic resonance imaging (MRI) in high-risk populations, including women with the BRCA mutation, using an evidence-based approach.

Methods: The MEDLINE, PubMed, EBM Reviews, ACP Journal Club, Cochrane Database MEDSEARCH, and SCOPUS databases were accessed and searched for articles up to August 2007. Articles were collected using the following terms and medical subject headings (MeSH) that applied to the focused clinical question: "BRCA1" and "BRCA2" with "mammography," "MRI," "prevention," "screening," and "surveillance." References from retrieved articles were also used to identify relevant papers. Abstracts were screened and relevant papers retrieved. Retrieved papers were graded for quality. Summary performance measures were obtained by random effects modeling of study-specific performance estimates and standard errors derived from the multiple 2 x 2 tables. Additionally, studies meeting the Centre for Evidence-Based Medicine level 2b quality were reviewed.

Results: In women with an increased risk without the BRCA gene, cancer detection rates by MRI were 0.011 (95% confidence interval [CI] 0.003-0.019), by mammography 0.005 (95% CI 0.002-0.008), and by a combination of both, 0.012 (95% CI 0.004-0.020). False-positive rates by MRI, mammography, or a combination of both were 0.10 (95% CI 0.03-0.18), 0.05 (95% CI 0.03-0.06), and 0.14 (95% CI 0.04-0.24). In BRCA positive women, cancer detection rates by MRI were 0.027 (95% CI 0.015-0.040), by mammography 0.010 (95% CI 0.005-0.016), and by a combination of both 0.031 (95% CI 0.018-0.045). False-positive rates by MRI, mammography, or a combination of both were 0.10 (95% CI 0.01-0.19), 0.05 (95% CI 0.03-0.07), and 0.14 (95% CI 0.04-0.24), respectively.

Conclusions: The data support an essential role for screening MRI in women with an increased risk for breast cancer.

Publication types

  • Meta-Analysis

MeSH terms

  • BRCA1 Protein / genetics*
  • BRCA2 Protein / genetics*
  • Breast Neoplasms* / diagnosis
  • Breast Neoplasms* / epidemiology
  • Breast Neoplasms* / genetics
  • Clinical Trials as Topic / statistics & numerical data
  • Evidence-Based Medicine / statistics & numerical data
  • Genetic Predisposition to Disease / epidemiology
  • Genetic Predisposition to Disease / genetics
  • Magnetic Resonance Imaging / statistics & numerical data*
  • Mammography / statistics & numerical data*
  • Mass Screening / methods*
  • Population Surveillance / methods
  • Prevalence
  • Reproducibility of Results
  • Risk Assessment / methods*
  • Risk Factors
  • Sensitivity and Specificity


  • BRCA1 Protein
  • BRCA2 Protein