Introduction

It is clinically desirable¹ to know the location and distribution of atherosclerotic plaque assists in preprocedural planning² and the detection of systemic atherosclerotic plaques at different arteries, such as the carotid arteries and aortic arch.
Direct thrombus imaging, based on magnetization prepared rapid gradient echo (MPRAGE), is one of the most frequently-used methods for evaluating the atherosclerotic plaque³. Recently, a new 3D MPRAGE direct thrombus imaging technique with a large field-of-view (FOV), from aortic artery to internal carotid artery, enables sufficient blood suppression and robust fat suppression using improved DIXON-based gradient echo sequence (mDIXON-MPRAGE)⁴. Furthermore, simultaneous acquisition of MR angiography (vessel burden information) and thrombus or intraplaque hemorrhage (IPH) detection⁵ could be helpful because it is time-efficient without motion-related misregistration.
On the other hand, a new non-contrast-enhanced, relaxation-based, flow-independent MRA method, called Relaxation-Enhanced Angiography without Contrast and Triggering (REACT) has recently been proposed for vascular imaging in various body parts with large anatomical coverage⁶. REACT consists of a 3D dual-echo Dixon segmented gradient echo and magnetization preparation using a T2prep pulse and a non-selective inversion recovery (IR) pre-pulse. We hypothesised that REACT can be extended to obtain MPRAGE images in one single scan because the applied IR pulse and the sequence structure is similar to mDIXON-MPRAGE⁴. In this study, we attempted to develop simultaneous MRA and MPRAGE imaging with a large FOV using REACT with Multiple Delays (REACT-MD).

Methods

REACT-MD sequence:
A schematic overview of the REACT-MD sequence is shown in Figure 1. It basically consists of two signal readouts by 3D Dixon dual-echo turbo field-echo (TFE) preceded by a T2prep pulse and a non-selective IR pulse [Fig.1a]. Magnetization preparation pulses were implemented to suppress signal from tissues such as muscles, nerves and organs, according to their difference in relaxation times. In detail, immediately after the T2prep pulse, an IR pulse was applied with a short inversion time (TI) to suppress tissues with short-to-intermediate T₁ and T₂, thus enhancing the blood-to-tissue contrast during the first signal readout for MRA [Fig.1b]. Thereafter, at the time of the blood null point, the second signal readout was carried out additionally to obtain the MPRAGE images [Fig.1a] . Consequently, REACT-MD provides MRA (phase-1) and MPRAGE (phase-2) images in one single scan without prolonging the total acquisition time. Besides, for robustness against B₁ and B₀ sensitivity across the entire the large-FOV, we applied a broadband offset independent trapezoid (OIT) pulse for both four T2prep refocusing pulses and the IR pulse, as already introduced for brachial plexus MR neurography and spine MR bone imaging^7-10.
Experiments:
A total of 5 volunteers and 5 patients were examined on a 3.0T whole-body clinical system (Ingenia, Philips Healthcare). The study was approved by the local IRB, and written informed consent was obtained from all subjects.
(1) Parameter optimization:
To investigate an optimal inversion delay (TFE phase interval, Fig.1a) for nulling the blood signal, we targeted for an optimal TI ranging from 500 to 1000ms with 50ms increments. To quantitatively compare the effect of respective inversion times, we measured the contrast-ratio (CR) between the vessel lumen and background muscles. First, circular-shaped regions-of-interest (ROIs) were placed over the ascending aorta, bilateral common carotid arteries and bilateral subclavian arteries on the REACT MRA images [Fig.2a]. Subsequently, these ROIs were copied to the same location over the REACT-MPRAGE with respective TI images. In addition, the ROIs for muscles were also placed over the bilateral sternocleidomastoid muscles [Fig.2b]. Finally, the CR values were calculated by using average signals of all vessels and the bilateral sternocleidomastoid muscles. Furthermore, images of respective TIs were visually assessed and compared by the radiologists and technologists.
(2) Comparison with conventional sequences:
To investigate the effect of REACT-MD, we visually compared the image quality with conventional REACT and mDIXON-MPRAGE in regards to overall SNR and the presence of artifacts.
Imaging parameters for REACT-MD were; Coronal, voxel size=1.00x1.00 x1.00mm³, TFE shot-interval=4000ms, TR=4.3ms, TE1/TE2=1.52/2.8ms, flip angle=12°, NSA=1, TFE factor=50, Compressed SENSE-reduction factor=3.3, and total acquisition time=5m0s.

Results and Discussion

The optimal TI was found to be around 600ms, when CR between blood and muscle reached maximum [Fig.2c]. The quality of REACT-MD MRA and MPRAGE images was considered comparable to that of the conventional REACT MRA and mDIXON-MPRAGE, respectively. Representative images from all these techniques are shown in Figure 3. It is noteworthy that REACT-MD provided MRA and MPRAGE images in one single scan and allowed for imaging over a large FOV at 3.0T. Two clinical cases with patients with systemic atherosclerotic plaques [Fig.4] and left subclavian artery occlusion due to the presence of thrombus [Fig.5] demonstrated that multiple plaques with IPH may be clearly visualized in REACT-MD with complementary MRA on vascular morphology. The clinical information provided was at least equivalent to that obtained from the conventional sequences separately. It may even be superior owing to uniform background tissue suppression at multiple delays, the efficacy of which would need to be investigated more systematically.

Conclusion

REACT-MD enables both MR angiography and MPRAGE direct thrombus imaging simultaneously, with uniform fat suppression over a large FOV in one single scan. It holds promise for assessment of systemic atherosclerosis and further clinical studies in comparison to the other state-of-the-art methods⁵ are warranted.

Acknowledgements

No acknowledgement found.