Introduction

Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) of the liver is typically performed using a breath-holding 3D gradient echo sequence. However, breath-holding for >10 s is sometimes difficult for some patients. In such patients, image quality may be impaired with respiratory motion artifacts during breath-hold acquisition, leading to diagnostic inaccuracy. Multiple arterial phase acquisition can be a solution for respiratory artifacts. Differential subsampling with Cartesian ordering (DISCO) is a high spatiotemporal-resolution dynamic contrast-enhanced MRI technique that combines a dual-echo SPGR sequence with pseudo-random variable-density k-space segmentation and view-sharing reconstruction¹. It could provide at least one image set with reduced or no artifacts. Another strategy for reducing the respiratory artifact is breath-hold-free acquisition. Respiratory gating with navigator echoes is available to obtain 3D gradient echo sequences without breath-holding. However, respiratory gating prolongs the acquisition time, leading to limited time resolution. DISCO-Star is a stack-of-stars technique that is robust for respiratory motion compared with Cartesian sampling² and is useful for obtaining free-breathing DCE-MRI. To date, there has been no direct comparison of the image quality of arterial phase (AP) imaging between DISCO and DISCO-Star. Thus, this study aimed to investigate the utility of AP imaging using DISCO-Star compared to DISCO.

Methods

This retrospective study was approved by the institutional review board, and the requirement for informed patient consent was waived. Forty-one patients who underwent gadoxetate disodium-enhanced MRI using DISCO and DISCO-Star at different time points with a 3-T MRI system were enrolled. Gadoxetate disodium (0.025 mmol/kg body weight) was administered at a rate of 1 mL/s, followed by a 20-mL saline flush using a power injector. Imaging parameters of DISCO and DISCO-Star are shown in Fig.1. In the DISCO-Star sequence, a total of 1500 radial spokes were acquired in 4 min continuously, and for routine frame rate reconstruction, 20 phases with 150 spokes per phase were reconstructed using 75 spokes offset between phases, which was equivalent to 12 s/phase. In each phase reconstruction, an auto-calibrating reconstruction for the Cartesian sampling (ARC) algorithm was used before Fourier transformation in the kz direction and conjugate gradient-sensitivity encoding (CG-SENSE) algorithm was used for in-plane reconstruction³. Soft-gating, a retrospective gating technique, was used to suppress image blur due to respiratory motion⁴. Visual assessment of the scan timing of AP (adequate or inadequate) was performed for each group, and in case of inadequate scan timing in the DISCO-Star group, retrospective reconstruction with a high frame rate (80 phases, 3 s/phase) was added. Adequate scan timing of AP was defined as that time when the antegrade flow enhanced the hepatic artery and portal vein but not the hepatic vein (Fig. 2). Artifact (5-point visual score: 5, no; 4, mild; 3, moderate; 2, extreme; 1, unreadable), overall image quality (5-point visual score: 5, excellent; 4, good; 3, acceptable; 2, poor; 1, non-acceptable), and the diagnosable image quality (≥3) were also assessed for each group and compared using the chi-square test. The non-inferiority of the AP image quality of DISCO-Star was compared to that of DISCO using the two one-sided McNemar tests with 5% non-inferiority margins.

Results

Inadequate scan timing of AP was observed in two patients (4.9%, 2/41) in the DISCO dataset and nine patients (22.0%, 9/41) in the DISCO-Star dataset, respectively. Adequate scan timing of AP was observed in the 3rd phase (21.2%, 7/32), 4th phase (66.7%, 22/32), or 5th phase (9.5%, 3/32) in the DISCO-Star dataset. (Fig. 3). In all patients with inadequate timing of AP in DISCO-Star dataset, retrospective additional reconstruction with a high frame rate offered images of adequate scan timing of AP (Fig. 4). DISCO-Star showed lower scores for streak artifacts and overall image quality (P < 0.0001). However, non-inferiority in the proportion of diagnosable images in the DISCO-Star dataset in comparison with the DISCO dataset (P = 0.036). (Fig. 5).

Discussion

Our results revealed that DISCO-Star showed a high performance in obtaining AP images with adequate scan timing. Although routine frame rate reconstruction (12 s/phase) was not perfect, retrospective high frame rate reconstruction (3 s/phase) salvaged the AP images in cases with inadequate scan timing of AP by routine frame rate. AP is the most important phase in DCE-MRI of the liver. There are some drawbacks to obtaining optimal AP images on gadoxetate disodium-enhanced MRI. Previous reports revealed that a high incidence (7%–14%) of patients self-reported transient dyspnea after the administration of gadoxetate disodium^5,6. Its injection volume is half that of extracellular contrast agents, which causes a short duration of adequate scan timing and truncation artifacts due to k-space inhomogeneity^7,8. DISCO-Star without breath-holding is a good solution to obtain optimal AP images because there is no worry of missing scan timing of AP and it is not affected by respiratory motion artifacts. Our results revealed that there was no significant difference in diagnosable image quality between the DISCO-Star and DISCO datasets. However, the degradation of image quality due to streak artifacts is a problem to be solved.

Conclusion

The use of DISCO-Star is the good solution to obtain the optimal AP images.

Acknowledgements

No acknowledgement found.

References

1. Saranathan M, Rettmann DW, Hargreaves BA, et al. DIfferential Subsampling with Cartesian Ordering (DISCO): a high spatio-temporal resolution Dixon imaging sequence for multiphasic contrast enhanced abdominal imaging. J Magn Reson Imaging. 2012;35:1484–1492. 2. Peters DC, Korosec FR, Grist TM, et al. Undersampled projection reconstruction applied to MR angiography Magn Reson Med. 2000;43:91-101. 3. Pruessmann KP, Weiger M, Bornert P, et al. Advances in sensitivity encoding with arbitrary k-space trajectories Magn Reson Med. 2001;46:638-651. 4. Johnson KM, Block WF, Reeder SB et al. Improved least squares MR image reconstruction using estimates of k-Space data consistency. Magn Reson Med. 2012;67:1600-1608. 5. Motosugi U, Bannas P, Bookwalter CA, et al. An investigation of transient severe motion related to gadoxetic acid-enhanced MR imaging. Radiology. 2016;279(1):93-102. 6. McClellan TR, Motosugi U, Middleton MS, et al. Intravenous gadoxetate disodium administration reduces breath-holding capacity in the hepatic arterial phase: A multi-center randomized placebo-controlled trial. Radiology. 2017;282(2):361-368. 7. Haradome H, Grazioli L, Tsunoo M, et al. Can MR fluoroscopic triggering technique and slow rate injection provide appropriate arterial phase images with reducing artifacts on gadoxetic acid-DTPA (Gd-EOB-DTPA)-enhanced hepatic MR imaging? J Magn Reson Imaging. 2010;32(2):334-340. 8. Motosugi U, Ichikawa T, Sou H, et al. Dilution method of gadolinium ethoxybenzyl diethylenetriaminepentaacetic acid (Gd-EOB-DTPA)-enhanced magnetic resonance imaging (MRI). J Magn Reson Imaging. 2009;30(4):849-854.