Evaluation of the role of dynamic contrast-enhanced MR imaging for patients with BI-RADS 3-4 microcalcifications

PLoS One. 2014 Jun 13;9(6):e99669. doi: 10.1371/journal.pone.0099669. eCollection 2014.

Abstract

Objective: The purpose of study was to prospectively evaluate the diagnostic performance of dynamic contrast-enhanced MR imaging in the differentiation of malignant lesions from benign ones in patients with BI-RADS 3-4 microcalcifications detected by mammography.

Materials and methods: 93 women with 100 microcalcifications had undergone breast MRI from June 2010 to July 2013. Subsequently, 91 received open biopsy and 2 received stereotactic vacuum-assisted biopsy. All results were compared with histological findings. The PPV, NPV and area under curve (AUC) of the mammography and breast MRI were calculated.

Results: There were 31 (31.0%) BI-RADS 3 microcalcifications and 69 (69.0%) BI-RADS 4. The PPV and NPV of mammography is 65.2% (45/69) and 90.3% (28/31). The PPV and NPV of breast MRI was 90.2% (46/51) and 95.9% (47/49). Among 31 BI-RADS 3 microcalcifications, the PPV and NPV of breast MRI was 100% (3/3) and 100% (28/28). Among 69 BI-RADS 4 microcalcifications, the PPV and NPV of breast MRI was 89.6% (43/48) and 90.5% (19/21). The AUC of mammography and breast MRI assessment were 0.738 (95% CI, 0.639-0.837) and 0.931 (95% CI, 0.874-0.988) (p<0.05).

Conclusion: Dynamic contrast-enhanced MR imaging of breast is able to be applied to predict the risk of malignance before follow-up for BI-RADS 3 microcalcifications and biopsy for BI-RADS 4 microcalcifications.

MeSH terms

  • Adult
  • Aged
  • Area Under Curve
  • Biopsy
  • Breast Neoplasms / diagnosis*
  • Calcinosis / diagnostic imaging
  • Calcinosis / pathology*
  • Female
  • Humans
  • Image Enhancement / instrumentation
  • Magnetic Resonance Imaging / instrumentation
  • Magnetic Resonance Imaging / methods*
  • Mammography / methods*
  • Middle Aged
  • Predictive Value of Tests
  • Prospective Studies
  • Radionuclide Imaging

Grants and funding

The authors have no funding or support to report.