Introduction

The static-fluid MRU is clinically useful for quickly and directly identifying the urinary tract dilation and obstruction.¹ Three-dimensional (3D) MRU could obtain contiguous thin-slice images allowing multi-plane reconstructions in any direction, and hence, providing better anatomic depiction and lesion conspicuity compared with 2D sequences. For conventional 3D respiratory-triggered fast spin-echo (RT-FSE) MRU, the most criticized point is its extra-long acquisition time. In contrast, the gradient and spin-echo (GRASE) sequence, with shorter acquisition time, could make a 3D MRU in a single breath-hold. The purpose of this study was to evaluate the performance of breath-hold GRASE (BH-GRASE) in 3D MRU.

Methods

From January to May 2021, 90 patients (47 men and 43 women; mean age, 49.3±19.1 years; range, 18-87 years) were prospectively included, and underwent both 3D BH-GRASE and RT-FSE MRU at 3.0 T (uMR 790, United Imaging Healthcare, Shanghai, China). Inclusion criteria were as follows: 1) no contraindications for MRI examination, 2) completion of two types of MRU, 3) no history of urinary tract surgery, 4) visible urinary tract dilation.
MR protocol for 3D BH-GRASE MRU parameters was: TR/TE=1300/ 267ms; parallel acquisition factor=3; FSE factor=35; EPI factor=7; FOV=400×400mm²; resolution (reconstruction) =1.88×1.32×3 (1.25×0.88×1.5) mm³; slice number=28; acquisition time=14.8s. For 3D RT-FSE MRU parameters: TR/TE=variable depending on respiratory/697 ms; parallel acquisition factor=2.7; FSE factor=205; FOV=400×400mm²; resolution (reconstruction) =1.42×1.14×2 (0.95×0.76×1) mm³; slice number=42; acquisition time=variable (range: 132–392s).
Three readers blinded to the MR protocols reviewed the source and MIP images. Reader A and B qualitatively evaluated the image quality using a five-point scale, including the overall image quality, artifact, urinary tract visualization. Reader B and C assessed the diagnostic performance related to urinary tract dilation, based on the degree of dilation, obstructive location, and imaging features (benign or malignant) at the obstructive site. The relative contrast ratio (CR) between the urinary tract and adjacent area was measured quantitatively by Reader C.
CR=(SI_U-SI_A) / (SI_U +SI_A)
where SI_U and SI_A stand for the signal intensities of the urinary tract and adjacent area, respectively. The acquisition time of two MRU protocols was recorded by Reader C. The reference standard was confirmed by the clinical data and the pathological results of the endoscope biopsy or surgery.

Results

The reference standard for the diagnosis of urinary tract pathology was based on subsequent surgery (n=22), endoscope (n=22). Of these, 21 were diagnosed with malignant lesion and 23 had benign stricture. The remaining 46 patients were clinically diagnosed due to benign consideration based on clinical information.
Overall, no significant difference were found for image quality between two MRU protocols. 3D BH-GRASE MRU better depicted bilateral renal calyces than RT-FSE MRU (p<0.05), Table 1.
For the quantitative evaluation of urinary tract contrast, CR values were slightly lower on BH-GRASE MRU (CR_pelvis=0.81±0.13; CR_ureter=0.91±0.12; CR_bladder=0.93±0.03) compared with RT-FSE MRU (CR_pelvis=0.86±0.09; CR_ureter=0.95±0.04; CR_bladder=0.96±0.02; all p values<0.05).
There were excellent agreements in the assessment of urinary tract dilation between two MRU protocols, including the dilated degree, the obstructive level, and the obstructive imaging features (inter-sequence κ=0.924-1). 3D BH-GRASE MRU achieved comparable diagnostic ability for diagnosing the etiology of obstruction with RT-FSE MRU (Fig. 1,2, Table 2).

Discussion

In this study, the performance of two different MRU methods, based on BH GRASE and RT-FSE respectively, was compared qualitatively and quantitatively. 3D MRU was successfully performed within a single breath-hold using the GRASE technique.
The GRASE sequence, proposed in the early 1990s, is a combination of spin- and gradient-echo sequence.² It provides shorter acquisition time than conventional FSE, due to the interleaving of the EPI readout between two consecutive 180 pulse (Fig. 3). In this study, 3D MRU was accelerated by adopting a FSE factor of 35, an EPI factor of 7, and a SENSE factor of 3. Acceleration of 3D MRU not only increases patient tolerance, but also decreases the motion-related artifacts of renal calyces.
As the multi-parametric MR urography requires a coverage of the entire urinary tract from abdomen to pelvic with several types of sequences,³ it is usually time-consuming, about 30 to 40min. As the 3D BH-GRASE MRU is quick and easy to perform, adding it into the routine multi-parametric MR urography protocol may provide not only a broad overview of the urinary tract, but also more diagnostic information.

Conclusion

Our study demonstrates the clinical feasibility of BH-GRASE in 3D static-fluid MRU for depicting the urinary tract obstruction and dilation, as it could be performed in a single a breath-hold without sacrificing image quality and diagnostic performance. BH-GRASE sequence could be a preferred clinically feasible protocol for 3D MRU.

Acknowledgements

We thank Yanqun Teng and Shuheng Zhang of United Imaging Healthcare for the technical support.