Diffusion weighted imaging in breast MRI: comparison of two different pulse sequences

Acad Radiol. 2007 Sep;14(9):1077-83. doi: 10.1016/j.acra.2007.06.006.

Abstract

Rationale and objectives: Comparison of two different diffusion weighted (DW) sequences in breast MRI regarding the differentiation between benign and malignant lesions.

Materials and methods: Breast MRI including two different DW sequences was performed in 165 consecutive women. Inclusion criteria for DW imaging and ADC evaluation were histologically proven focal mass lesions with a diameter of more than 5 mm in dynamic contrast-enhanced MRI. The DW sequences were pre-contrast echo-planar imaging with spectral fat saturation (EPI fs) and DW EPI with inversion recovery (EPI STIR) (b-values: 50, 400, and 800). Lesions were analyzed regarding visibility in DW sequences and ADC values.

Results: Inclusion criteria were fulfilled in 56 women with 69 lesions. Five lesions could not be evaluated for different reasons. Finally, DW sequences were evaluated in 51 women with 64 focal mass lesions (15 benign, 49 malignant). The visibility of the lesions was significantly better in the EPI fs sequence (P<0.05). The ADC values (10(-3) mm(2)/s) in the EPI fs were 1.76, 2.58, and 1.21 (mean, maximum, minimum, respectively) for benign lesions and 0.90, 1.19, and 0.34 for malignant lesions. Respective values in the EPI STIR sequence were 1.92, 3.20, 1.10, and 0.91, 1.43, 0.35. Only in the EPI fs sequence there was no overlap in ADC values between benign and malignant lesions.

Conclusion: The DW MRI of the breast with EPI fs and EPI STIR sequences has a high potential to differentiate between benign and malignant breast lesions. Due to better lesion visibility and selectivity, the EPI fs sequence should be preferred.

Publication types

  • Comparative Study

MeSH terms

  • Breast Neoplasms / diagnosis*
  • Diffusion Magnetic Resonance Imaging / methods*
  • Echo-Planar Imaging / methods*
  • Female
  • Humans
  • Image Interpretation, Computer-Assisted / methods*
  • Middle Aged
  • Reproducibility of Results
  • Sensitivity and Specificity
  • Signal Processing, Computer-Assisted