Elsevier

Clinical Radiology

Volume 70, Issue 3, March 2015, Pages 286-294
Clinical Radiology

Breast DWI at 3 T: influence of the fat-suppression technique on image quality and diagnostic performance

https://doi.org/10.1016/j.crad.2014.11.012Get rights and content

Aim

To evaluate two fat-suppression techniques: short tau inversion recovery (STIR) and spectral adiabatic inversion recovery (SPAIR) regarding image quality and diagnostic performance in diffusion-weighted imaging (DWI) of breast lesions at 3 T.

Materials and methods

Ninety-two women (mean age 48 ± 12.1 years; range 21–78 years) underwent breast MRI. Two DWI pulse sequences, with b-values (50 and 1000 s/mm2) were performed with STIR and SPAIR. The signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), suppression homogeneity, and apparent diffusion coefficient (ADC) values were quantitatively assessed for each technique. Values were compared between techniques and lesion type. Receiver operating characteristics (ROC) analysis was used to evaluate lesion discrimination.

Results

One hundred and fourteen lesions were analysed (40 benign and 74 malignant). SNR and CNR were significantly higher for DWI-SPAIR; fat-suppression uniformity was better for DWI-STIR (p < 1 × 10−4). ADC values for benign and malignant lesions and normal tissue were 1.92 × 10−3, 1.18 × 10−3, 1.86 × 10−3 s/mm2 for DWI-STIR and 1.80 × 10−3, 1.11 × 10−3, 1.79 × 10−3 s/mm2 for SPAIR, respectively. Comparison between fat-suppression techniques showed significant differences in mean ADC values for benign (p = 0.013) and malignant lesions (p = 0.001). DWI-STIR and -SPAIR ADC cut-offs were 1.42 × 10−3 and 1.46 × 10−3 s/mm2, respectively. Diagnostic performance for DWI-STIR versus SPAIR was: accuracy (81.6 versus 83.3%), area under curve (87.7 versus 89.2%), sensitivity (79.7 versus 85.1%), and specificity (85 versus 80%). Positive predictive value was similar.

Conclusion

The fat-saturation technique used in the present study may influence image quality and ADC quantification. Nevertheless, STIR and SPAIR techniques showed similar diagnostic performances, and therefore, both are suitable for use in clinical practice.

Introduction

The role of diffusion-weighted MRI (DWI) in the characterization and differentiation of breast lesions remains a subject of intensive research.1, 2, 3 In breast DWI, the use of an effective fat-suppression technique is especially critical. As the MRI signal has contributions from both water and fat components,4 it is essential to efficiently eliminate the lipid signal. The fat fraction in breast tissue can be high, and as the apparent diffusion coefficient (ADC) of fat is much lower than that of water within lesions and/or normal glandular tissue,5 the presence of lipids may compromise an accurate ADC estimate.

Multiple fat-suppression techniques are currently available [short tau inversion recovery (STIR), spectral adiabatic inversion recovery (SPAIR), frequency-selective fat saturation (FatSat), water-selective excitation or the Dixon technique]. Each one is based on different physical phenomena,4, 5, 6 and previous studies have described their variable efficiency in lesion detection, and ADC estimates.7, 8, 9, 10

Given that the available techniques are not equally robust to the static magnetic field (B0) and radiofrequency magnetic field (B1) inhomogeneities, it is important to compare their performance on breast DWI at 3 T. Previous studies7, 8, 11 have used different techniques for fat suppression: Bogner et al.7 used STIR, whereas Peters et al.8 and El Khouli et al.11 used SPAIR, which makes direct comparisons difficult. For example, comparing results previously obtained by the present authors' group using a SPAIR-based sequence12 with those reported by Bogner et al.7 using STIR (b-values 50 and 1000 s/mm2 in both cases), similar ADC values were obtained for malignant and normal glandular tissue, but not for benign lesions, which could potentially be related to differences in fat-suppression efficiency.

Echo planar imaging (EPI) is a fast acquisition technique commonly used in breast DWI. It enables high imaging speed at the cost of being prone to chemical shift artefacts and geometric distortions.13 A common strategy to decrease these artefacts is to use parallel imaging (PI), an image reconstruction technique that makes use of sensitivity differences between different coil channels to perform spatial encoding, enabling a reduction in the number of phase-encoding steps. This in turn leads to a shorter readout window and hence reduced geometric distortion in the images.14 Another study developed by the present authors comparing DWI-STIR and -SPAIR at 3 T including PI15 [in press] revealed similar contrast-to-noise ratio (CNR) and signal-to-noise ratio (SNR) except for benign lesions, and comparable ADC values for benign and malignant lesions.

However, some authors reported a decrease in SNR and CNR with PI as the echo time (TE) shortening achievable with PI can be insufficient to compensate for the g-factor noise penalty, which depends on the coil geometry and reflects the difference in sensitivity of the available channels along the phase encode direction.16, 17 Given that the present DWI sequence includes higher b-values (b = 2000 and 3000 s/mm2) the decision was made to exclude PI to gain SNR. Thus, the purpose of the present study was to compare quantitatively DWI-STIR and -SPAIR when no PI is used regarding SNR and CNR, fat-suppression uniformity, and ADC quantification for lesion differentiation and characterization in the clinical setting.

Section snippets

Patients and lesions

The present study is included in a wider investigation focusing on the application of DWI to study breast lesions, for which approval has been obtained from the institutional review board (code CES 276/13). This prospective study was performed on women with clinical indication for breast MRI. Written informed consent was obtained from all patients.

Women were excluded from this study if they (1) had undergone chemotherapy or radiotherapy 24 months prior to the MRI examination (three women with

Patients and lesion characteristics

Ninety-two women (mean ± standard deviation age of 48 ± 12 years; range 21–78 years) with 114 lesions were successfully scanned with DWI-STIR and -SPAIR. Thirty-seven women were post-menopausal. One hundred and four lesions were mass (91.2%) and 10 non-mass lesions (8.8%). Among the 114 lesions, 74 were classified as malignant and 40 as benign. Histological results were obtained for 99 lesions by biopsy and/or surgery. Mean size for benign lesions was 13.±9 mm, whereas for malignant lesions it

Discussion

Achieving adequate elimination of the lipid signal is challenging, especially at magnetic field strengths ≥3 T due to increased susceptibility artefacts, image ghosting, and larger chemical shifts that can be present when compared to 1.5 T.14 To address these issues while preserving image quality, the use of an adequate fat-suppression method is essential to mitigate image artefacts and promote adequate SNR and CNR.24 Also, unsuppressed fat signal interferes with the diffusivity characteristics

Conclusion

  • The fat saturation technique used in breast DWI influences image quality.

  • Fat suppression uniformity was better for DWI-STIR than -SPAIR.

  • ADC quantification and cut-off values depend on the fat suppression used.

  • Diagnostic performance between DWI-STIR and -SPAIR was not significantly different.

  • DWI-SPAIR seems to be more accurate using as reference histological results.

Acknowledgments

The authors thank to the Ethics Committee of Hospital São João/Faculty of Medicine of Porto University (FMUP). This work was sponsored by Foundation of Science and Technology /Polytechnic Institute of Porto (grant number: SFRH/BD/50027/2009) and (grant number: PEst-OE/SAU/UI0645/2011).

References (40)

  • G.G. Lo et al.

    Diffusion-weighted magnetic resonance imaging of breast lesions: first experiences at 3 T

    J Comput Assist Tomogr

    (2009)
  • T.A. Bley et al.

    Fat and water magnetic resonance imaging

    J Magn Reson Imaging

    (2010)
  • J. Ma

    Dixon techniques for water and fat imaging

    J Magn Reson Imaging

    (2008)
  • W. Bogner et al.

    Diffusion-weighted MR for differentiation of breast lesions at 3.0T: how does selection of diffusion protocols affect diagnosis?

    Radiology

    (2009)
  • N.H. Peters et al.

    Quantitative diffusion weighted imaging for differentiation of benign and malignant breast lesions: the influence of the choice of b-values

    J Magn Reson Imaging

    (2010)
  • A. Matsuoka et al.

    Comparison of 3.0-and 1.5-Tesla diffusion-weighted imaging in the visibility of breast cancer

    Radiat Med

    (2008)
  • R.H. El Khouli et al.

    Diffusion-weighted imaging improves the diagnostic accuracy of conventional 3.0-T breast MR imaging

    Radiology

    (2010)
  • L. Nogueira et al.

    Diffusion-weighted imaging: determination of the best pair of b-values to discriminate breast lesions

    Br J Radiol

    (2014)
  • S.T. Skare et al.

    EPI-based pulse sequences for diffusion tensor MRI

  • R. Rakow-Penner et al.

    Breast MRI at 3T

    Appl Radiol

    (2009)
  • Cited by (29)

    • Tumor grading of soft tissue sarcomas: Assessment with whole-tumor histogram analysis of apparent diffusion coefficient

      2022, European Journal of Radiology
      Citation Excerpt :

      Therefore, the inclusion of lipid component within a ROI results in a low ADC, which may lead to misinterpretation as hypercellular regions [31]. It is known that even when fat suppression techniques are applied to DWI, insufficient fat suppression may result in a low ADC, especially with spectral adiabatic inversion recovery fat suppression [32]. Spectral adiabatic inversion recovery is based on spectrally selective pulses that only suppress the signal from the largest lipid peak which resonates at a frequency of about 3.5 ppm relative to water, and minor lipid peaks at different resonance frequencies remain unsuppressed [33].

    • Diffusion weighted imaging for evaluation of breast lesions: Comparison between high b-value single-shot and routine readout-segmented sequences at 3 T

      2021, Magnetic Resonance Imaging
      Citation Excerpt :

      The ADC values of the rs-EPI are comparable to those in previously reported studies and, therefore, the ADC values obtained with the ss-EPI-mIRBS were relatively low [19,20]. In fact, they were even low when compared to studies that used ss-EPI in combination with extremely high b-values [19–21–22]. There are several explanations that might cause this difference.

    View all citing articles on Scopus
    View full text