Original contribution
Improvement of renal diffusion-weighted magnetic resonance imaging with readout-segmented echo-planar imaging at 3 T

https://doi.org/10.1016/j.mri.2015.02.023Get rights and content

Abstract

Purpose

To assess the feasibility of a respiratory-gated implementation of readout-segmented SE-EPI (RESOLVE) for renal diffusion-weighted imaging (DWI) by comparison with single-shot SE-EPI (ss-EPI) in a phantom, healthy volunteers and chronic kidney disease (CKD) patients.

Materials and Methods

A fluid-filled phantom, 20 healthy volunteers and 10 CKD patients were scanned with the same parameters and coils on a 3 T MR system with 3 DW sequences (b-values = 0, 300, 500, 900 s/mm2): a standard ss-EPI (Reference EPI), a ss-EPI with higher resolution, bandwidth and acceleration factor (HR-EPI) and RESOLVE with the same spatial resolution as HR-EPI but a segmentation of the readout into 5 shots. Geometric distortions, image blurring using a ‘Canny’ edge detection based measure, cortico-medullary differentiation measured on b0 images and ADC quantification were compared between the 3 sequences using one-way analysis of variance (ANOVA) with post-hoc Bonferroni (p < 0.05 was taken as statistically significant).

Results

RESOLVE reduced significantly geometric distortions and blurring and improved, in the volunteers and patients, the sharpness score by 56% on average in comparison to ss-EPI (p = 0.02). In healthy volunteers, the cortico-medullary differentiation with RESOLVE was also possible on a wider range of b-values (p < 0.02) with ADC values (in 10 6 mm2/s) of 1994 ± 246 in the cortex and 1762 ± 238 in the medulla (p < 0.001). In CKD patients, ADC values (in 10 6 mm2/s) from the RESOLVE sequence were not different between the cortex (1755 ± 145) and the medulla (1799 ± 163, p = 0.49).

Conclusion

Despite a longer scan time, RESOLVE enhanced significantly the quality of renal diffusion-weighted images by improving the difference in SI and ADC between the renal cortex and medulla in healthy volunteers. In CKD patients, RESOLVE showed a disappearance of this cortico-medullary ADC difference. These improvements justify further clinical studies.

Introduction

Diffusion-Weighted MR Imaging (DWI) has shown promising results to differentiate pathological from healthy tissues in renal tumors [1], transplant rejection [2], pyelonephritis [3], ureteral stone obstruction [4] and renal artery stenosis [5]. DWI techniques used in the abdomen rely on single-shot diffusion-weighted echo-planar imaging (ss-EPI) which is sensitive to image artifacts [6]. The trade-off between resolution and signal-to-noise ratio for the large FOV used in the abdomen imposes an increased matrix size and therefore a longer EPI readout time. The sequence becomes more sensitive to in-plane geometric distortions caused by the off-resonance water protons in areas where a significant difference in susceptibility exists. In renal DW applications, severe distortions are present at the bowel (filled with air) and tissue interface.

One solution to improve the distortions in diffusion MRI is to use a ‘Readout Segmentation Of Long Variable Echo-trains’ (RESOLVE) MR sequence in combination with parallel imaging technique, such as GRAPPA, GeneRalized Autocalibrating Partially Parallel Acquisitions. The RESOLVE strategy is based on a segmentation of k-space into several shots along the readout direction in order to shorten the echo spacing. RESOLVE combined with parallel imaging was previously introduced by Porter et al. for acquiring high-resolution DW images with low susceptibility based image distortion and T2 blurring in the brain [7]. This strategy has been validated in non-triggered applications, such as in head and breast imaging to reduce sensitivity to susceptibility artifacts [8], [9], [10]. RESOLVE outperformed ss-EPI for analysis of the pediatric brain in regions prone to geometric distortions, such as the orbit, the skull base, and the posterior fossa [11]. By comparison with a conventional ss-EPI sequence, RESOLVE has improved the lesion-to-background contrast and the categorization between benign and malignant breast lesions [12]. For these reasons, we hypothesized that RESOLVE could improve the robustness against artifacts that are a consequence of the long k-space sampling in renal DWI. We proposed in this study an implementation of a respiratory-gated RESOLVE protocol for kidney DWI as well as a comparison of image quality with a different implementation of ss-EPI in a phantom, healthy volunteers and chronic kidney disease (CKD) patients.

Section snippets

Subjects

Twenty healthy volunteers, comprising 11 females and 9 males, with a mean age of 28.8 ± 4.7 years (range, 23–39 years) and ten patients, comprising 4 females and 6 males, with a mean age of 55 ± 16 (ranges 26–78 years), were recruited after informed consent. Healthy volunteers enrolled in this study had no known kidney disease. All patients were chronic kidney disease (CKD) patients, comprising 1 native kidney and 9 allograft patients. The patients' characteristics are summarized in Table 1. The study

Geometric distortions

The level of geometric distortions in the phantom is shown in Fig. 1a. The strongest deformation of the phantom borders was observed with the reference EPI, followed by the HR-EPI as quantified by the deviation from the corner of the phantom on the undistorted GRE reference image. The RESOLVE images were characterized by the smallest geometrical deformation among all the sequences. The maximum distance in the phase-encoding direction between the phantom edges of the GRE image and the reference

Discussion

Our goal was to study the potential of readout-segmented echo-planar imaging with a fivefold segmented k-space acquisition for renal DWI. An improved image quality from the RESOLVE sequence over ss-EPI sequences was observed in both healthy volunteers and CKD patients. The respiratory-gated RESOLVE significantly reduced diffusion artifacts and the other hurdles encountered with the use of ss-EPI. Whereas such improvements have already been demonstrated in the brain and breast [8], [9], [10],

Conclusions

In conclusion, despite a longer scan time, RESOLVE enhanced significantly the quality of renal diffusion-weighted images by reducing the image distortion and blurring and by improving the difference in SI and ADC between the renal cortex and medulla in healthy volunteers. The performance of the RESOLVE was also demonstrated in CKD patients with a disappearance of the cortico-medullary ADC difference. These improvements justify further clinical studies of the potential of RESOLVE for diffusion

Acknowledgements

This work was supported by grants from the Clinical Research Center of the Medicine Faculty of Geneva University and Geneva University Hospital and the Louis-Jeantet Foundation (CRC) and in part by the Centre for Biomedical Imaging (CIBM) of EPFL, University of Geneva and the University Hospitals of Geneva and Lausanne, as well as the Leenaards and Insuleman foundation.

References (37)

  • H.C. Thoeny et al.

    Functional evaluation of transplanted kidneys with diffusion-weighted and BOLD MR imaging: initial experience

    Radiology

    (2006)
  • R. Faletti et al.

    Diffusion-weighted imaging and apparent diffusion coefficient values versus contrast-enhanced MR imaging in the identification and characterisation of acute pyelonephritis

    Eur Radiol

    (2013)
  • H.C. Thoeny et al.

    Noninvasive assessment of acute ureteral obstruction with diffusion-weighted MR imaging: a prospective study

    Radiology

    (2009)
  • D. Le Bihan et al.

    Artifacts and pitfalls in diffusion MRI

    J Magn Reson Imaging

    (2006)
  • D.A. Porter et al.

    High resolution diffusion-weighted imaging using readout-segmented echo-planar imaging, parallel imaging and a two-dimensional navigator-based reacquisition

    Magn Reson Med

    (2009)
  • S.J. Holdsworth et al.

    Clinical application of readout-segmented- echo-planar imaging for diffusion-weighted imaging in pediatric brain

    AJNR Am J Neuroradiol

    (2011)
  • W. Bogner et al.

    Readout-segmented echo-planar imaging improves the diagnostic performance of diffusion-weighted MR breast examinations at 3.0 T

    Radiology

    (2012)
  • M. Iima et al.

    Reduced-distortion diffusion MRI of the craniovertebral junction

    AJNR Am J Neuroradiol

    (2012)
  • Cited by (43)

    • Diffusion-magnetic resonance imaging predicts decline of kidney function in chronic kidney disease and in patients with a kidney allograft

      2022, Kidney International
      Citation Excerpt :

      The strength of both the static magnetic field and gradients has been well recognized as a key factor in the improvement of diffusion MRI. Furthermore, we used a RESOLVE diffusion sequence based on segmented acquisition rather than the single shot approach of Sugiyama et al., according to the best overall performance of this sequence for prostate and renal imaging.20,24 The RESOLVE sequence allows a much shorter echo train with reduced signal blurring due to T2∗ decay25 and yields better correlation of ADC and renal fibrosis than traditional single-shot echo-planar imaging MR sequences.21

    • Clinical comparison of single-shot EPI, readout-segmented EPI and TGSE-BLADE for diffusion-weighted imaging of cerebellopontine angle tumors on 3 tesla

      2021, Magnetic Resonance Imaging
      Citation Excerpt :

      Readout-segmented echo-planar DWI (RESOLVE-DWI) is an advanced DWI technique in which k-space is divided into several segments along the readout direction, resulting in reduced echo spacing and echo time. Thus, the image distortions, T2* blurring and susceptibility artifacts of RESOLVE-DWI are substantially reduced by a factor proportional to the segment width [11–14], and it has been widely applied in clinical imaging of brain, sacroiliitis, prostate, breast, kidney, and rectum [15–20]. However, RESOLVE-DWI is still vulnerable to susceptibility artifacts due to its due to its use of an intrinsic EPI readout scheme, and image distortions and T2* blurring effect are not completely eliminated [21,22].

    View all citing articles on Scopus
    View full text