2058
REACT - Relaxation-Enhanced MR Angiography without Contrast and Triggering in comparison with other non-contrast MR angiography techniques for imaging of the pelvic vessels.1Diagnostic and Interventional Radiology, University Hospital RWTH Aachen, Aachen, Germany, 2Philips Medical Systems, Best, Netherlands
Synopsis
Conventional MRI and CE-MRA have limitations in venous imaging. Recently introduced Relaxation-Enhanced MR Angiography without Contrast and Triggering (REACT) is based on 3D dual-echo generalized DIXON in combination of magnetization preparation, and has the potential to better characterize both arterial and venous signals with robust blood-tissue contrast. Comparing REACT with conventional non-contrast sequences regarding vessel depiction, REACT revealed similar results for large vessels and was superior for the assessment of small peripheral vessels.
Introduction
Contrast enhanced MRA is not suitable for body MR angiography of the venous system. Instead, several flow-independent MR techniques have been proposed for body venous vessel imaging. However, most of these techniques were not primarily designed for angiography. Thus an exact delineation of small and mid-size vessels is often challenging. Purpose of this study was to evaluate a new pulse sequence (REACT - Relaxation-Enhanced MR Angiography without Contrast and Triggering) for pelvic vessel imaging, and to compare its utility with that of bFFE/bSSFP and plain TSE vessel imaging.Methods
A total of five healthy volunteers and two patients were examined on a 1.5T MR-system (Philips Ingenia). Written informed consent was obtained from all subjects. All images were acquired in a transversal orientation using a 16-channel anterior torso coil array and a 16-channel posterior spine coil built into the patient table. Conventional T2-weighted turbo spin-echo, a balanced fast field-echo (bFFE/bSSFP) sequence and the REACT technique were applied at the pelvic level. REACT is based on a 3D-two-point generalized Dixon TFE sequence, with additional magnetization preparation by a non-volume selective adiabatic inversion pulse preceded by a four-refocusing-adiabatic-pulse T2-prep module [1]. While the latter collectively suppresses signals from static tissues, such as muscles and organs with short T1 and T2 values, the Dixon acquisition additionally suppresses the residual fat signal. As a result, signal of tissue with long T1 and long T2, i.e. blood signal, is enhanced, and thus an optimal vessel-to-background contrast [2, 3] is achieved. Imaging parameters are shown in Table 1.
Image quality was assessed on a 4-point scale, on five representative images on different pelvic levels within the three different pulse sequences, and rated with regards to the following tasks: (a) anatomic delineation of large proximal vessels, (b) delineation of small distal vessels, and (c) degree of image degradation by artefacts. To evaluate over- or undersizing of REACT compared to the conventional techniques, vessel diameters (abdominal aorta, inferior vena cava (IVC), both iliac arteries) were measured. Vessel diameters were compared using Friedman tests; image quality was assessed using Wilcoxon matched-pair signed-Rank tests.
Results
Image acquisition was successful in all subjects. Average image quality scores for the visibility of large vessels for T2-TSE, bSSFP and REACT was rated as 3.5, 3.3, and 3.7, respectively, resulting in a significant difference between REACT vs. bSSFP, but not between the other pairs. Average image quality scores for the visibility of small vessels increased from T2-TSE over SSFP to REACT as follows: 1.6, 2.2, 3.2, yielding statistically significant differences between the three sequences. Artefact degradation was lowest with T2-TSE (mean 3.9), whereas SSFP (3.2) and REACT (3.3.) had a similar artefact level. Vessel diameters as measured in T2-TSE vs. bSSFP vs. REACT were as follows: distal aorta 14.7 vs. 15.1 vs. 15.3 mm; inferior vena cava: 8.2 vs. 10.9 vs. 10.2 mm; left iliac arteries 9.3 vs. 9.2 vs. 9.1 mm; right iliac arteries: 9.3 vs. 9.5 vs. 9.7 mm. Except for the short axis of the IVC there were no statistical differences between the three sequences in these anatomical measurements of vessel diameters.Discussion
REACT allows for an accurate delineation of proximal and distal pelvic vessels without significant over-/undersizing. The significant differences between the IVC diameter on T2-TSE and REACT or bSSFP can be attributed to the lack of breath triggering in bSSFP und REACT. In particular, compared with the usual bSSFP pulse sequence, REACT significantly improved delineation of smaller peripheral vessels. Despite artefact degradation was similar in REACT vs. bSSFP, the improved imaging of small peripheral vessels was explained by an improved background suppression in REACT images. Although it was not explicitly assessed, this suppression led to poorer visibility of the surrounding pelvic anatomy similar to other MRA techniques.Conclusion
In particular in the peripheral vessels, REACT is superior to other conventional non-contrast enhanced techniques. Thus, REACT is especially suitable for the assessment of the venous system where the imaging using CE-MRA often is challenging.Acknowledgements
No acknowledgement found.References
1. Yoneyama M, et al. ISMRM 2016:2252.
2. Toyonari N, et al. ISMRM 2016:2684.
3. Pokorney AL, et al. ISMRM 2017:0335.
Figures
