Skip to main content

Value of diffusion-weighted imaging in differentiating benign from malignant breast lesions

Introduction

MRI has a proven high sensitivity for the detection of breast cancer. Besides contrast enhancement which is related to the vascularity of breast tumours, tumour high cellularity can be characterized or better measured via diffusion-weighted imaging (DWI), which is an important characteristic of tumours. DWI, combined with morphology and kinetics of contrast enhancement, may increase the specificity of MRI and may help in differentiating between normal and malignant breast lesions.

Materials and methods

So far, 11 patients with biopsy-proven breast cancer and eight patients with benign breast lesions underwent DWI of the breast with an echo-planar imaging (EPI) sequence at 1.5 T. The apparent diffusion coefficient (ADC) was measured using four different b-values: 0, 300, 500, and 800 seconds/mm2.

Results

The ADC values of the malignant lesions ranged from 0.82 × 10-3 mm2/seconds to 1.65 × 10-3 mm2/seconds, with a mean ADC value of 1.14 ± 0.22 × 10-3 mm2/seconds. In benign lesions, the mean ADC was 1.62 ± 0.24 × 10-3 mm2/seconds, varying from 1.17 × 10-3 mm2/seconds to 2.00 × 10-3 mm2/seconds.

Conclusion

The preliminary results show that contrast-enhancement patterns are mostly aspecific, while there is a better concordance between tumour malignancy and ADC values. Inclusion of more patients in the future might result in a threshold value of ADC allowing malignant lesions to be distinguished from benign lesions.

Author information

Affiliations

Authors

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Stijven, S., Gielen, E., Horvath, M. et al. Value of diffusion-weighted imaging in differentiating benign from malignant breast lesions. Breast Cancer Res 11, P16 (2009). https://doi.org/10.1186/bcr2299

Download citation

Keywords

  • Breast Cancer
  • Diffusion Coefficient
  • Cancer Research
  • Tumour Malignancy
  • Apparent Diffusion Coefficient