Giulia Ginami1, Imran Rashid1, Harith Alam1, Radhouene Neji1,2, Israel Valverde1,3, Alessandra Frigiola1, René Michael Botnar1, and Claudia Prieto1
1School of Biomedical Engineering and Imaging Science, King's College London, London, United Kingdom, 2MR Research Collaborations, Siemens Healthcare Limited, Frimely, United Kingdom, 3Hospital Universitario Virgen del Rocio, Seville, Spain
Synopsis
Bright- and black-blood MRI sequences provide complementary diagnostic information in patients with congenital heart disease (CHD). Typically, contrast agent administration is needed to depict structures such as the pulmonary veins or regions of disturbed blood flow. However, contrast-enhanced imaging is not ideal for regular screening, while the acquisition of bright- and black-blood sequences in a sequential fashion remains sub-optimal. Here we propose a motion-compensated 3D whole-heart sequence that provides co-registered bright- and black-blood volumes without the need for contrast agent administration. The sequence was tested in patients with CHD and showed high-quality delineation of both arterial and venous structures.
Purpose
Congenital heart disease (CHD) represents one of the most common types of birth defect. CHD can affect different cardiac structures, such as the aorta, the pulmonary veins, as well as the atrial and ventricular septum. Cardiovascular MRI plays an important role in both diagnosis and follow-up of patients with CHD. Bright-blood acquisitions are performed for the visualization of the great vessels and of the coronary course, while black-blood images are preferred for the depiction of small vessels. Contrast-agent administration is often required for the delineation of the pulmonary veins, the collateral vessels, or regions with perturbed blood flow. However, the reliance on intravenous administration of Gadolinium (Gd)-based contrast agents is not ideal, especially in such a patient population where repeated scans are required to monitor progresses. Furthermore, the sequential acquisition of bright- and black-blood images may entail prolonged and unpredictable examination times, thus limiting the achievable volumetric coverage and spatial resolution. In this study, we tested in CHD patients a novel motion-compensated 3D whole-heart sequence that provides co-registered bright- and black-blood volumes without the need for contrast agent administration.
Methods
The proposed prototype sequence resembles that in 1 but exploits magnetization transfer contrast (MTC) to preserve the signal from both arterial and venous blood while generating adequate contrast between the blood itself and the myocardium (Fig.1). Such MT prepared bright- and black-blood (BOOST) sequence has been originally introduced for the assessment of pulmonary vein anatomy in patients suffering from atrial fibrillation2, and it is here extended to the screening of patients with CHD. The sequence exploits a Cartesian trajectory with spiral profile ordering3 and alternates the acquisition of an MT-prepared inversion recovery (IR) pulse (MTC-IR BOOST, odd heartbeats) and that of MT-preparation solely (MTC BOOST, even heartbeats). MT prepared BOOST is integrated with image-based navigation4 and non-rigid respiratory motion correction5 to achieve 100% respiratory scan efficiency (no data rejection) and thus shorter and predictable scan times. After the motion-corrected reconstruction of the two differently weighted bright-blood volumes, those are combined in a phase-sensitive IR (PSIR) reconstruction6 to obtain a third, complementary, and fully co-registered black blood volume. Data acquisition: was performed in 12 patients with CHD (7 males, 28.4±13.1 years) on a 1.5T system (Siemens MAGNETOM Aera). Imaging parameters included: resolution 1.4x1.4x2.8 mm (reconstructed to 1.4mm3), flip-angle 90deg, TE/TR 1.5/3.2ms, MT-preparation: 15 Gaussian pulses, flip-angle 800deg, frequency offset 3000Hz, duration 20.5ms. Additionally, a standard T2-prepared (40ms) non-contrast enhanced (non-CE) 3D whole-heart bright-blood clinical sequence was acquired for comparison purposes (bSSFP readout, resolution 1.5mm3, acceleration GRAPPA 2x, flip-angle 90deg). Data analysis: The signal to noise of both arterial and venous blood (SNRart and SNRven) was computed for bright-blood MTC-IR BOOST dataset, while contrast to noise was computed for both the bright-blood MTC-IR BOOST and the clinical sequences (CNRart and CNRven). Diameters of the great vessels were assessed as per standardized clinical reporting for both bright-blood sequences.
Results
Bright- and black- blood MT prepared BOOST images are shown in Fig.2 for two representative patients. The bright-blood MTC-IR BOOST provided excellent delineation of the pulmonary veins, and showed improved luminal signal depiction in the case of altered blood flow (e.g in the context of pulmonary regurgitation/stenosis or coarctation) (Fig. 2 and 3). The black-blood PSIR BOOST dataset showed uniform and flow-independent blood signal suppression (Fig.2 and 3). Bright-blood MTC-IR BOOST and T2 prepared clinical sequences are compared in Fig. 4 for 3 patients. The bright-blood MTC-IR BOOST images provided high-quality depiction of both arterial and venous blood, with quantified SNRart=21.6±9.5 and SNRven=19.9±8.3, P=NS, and CNRart=12.7±3.2 and CNRven=10.9±2.5ms, P=NS. Conversely, and while providing high-quality delineation of the arterial structures (CNRart=13.7±5.6, P=NS with respect to BOOST), the conventional T2-prepared sequence lead to a degraded depiction of the venous blood (CNRven=5.5±4.3, P<0.005 with respect to CNRart). Furthermore, the two non-CE sequences – MTC-IR BOOST and clinical T2-prepared – showed excellent agreement for the measurement of the great vessels diameters (Fig.5).
Discussion and Conclusion
The MT-prepared BOOST sequence was successfully tested in a cohort of patients with CHD. It provided bright- and black-blood high-quality visualization of the cardiac anatomy, including both the arterial and the venous systems. Vein delineation significantly improved with BOOST when compared to a more standard non-CE clinical sequence. Therefore, BOOST holds promise for obviating the requirement of Gd usage in patients with CHD, making it an ideal candidate for screening and follow-up in this population. Future technical developments will aim at the integration of BOOST with acceleration techniques in order to improve both the spatial resolution and the acquisition times. Once those technical developments are implemented, the clinical validation of this sequence in a larger patient population is foreseen.
Acknowledgements
Grant sponsors: EPSRC EP/N009258/1, EP/P001009/1, EP/P007619/1,MRC MR/L009676/1, and FONDECYT N° 1161051. The views expressed are those of the authors and notnecessarily those of the National Health Service, theNational Institute for Health Research, or the Departmentof Health.
References
[1] Ginami G et al, Magn Reson Med. 2018; 79(3):1460-1472. [2] Ginami G et al, Magn Reson Med. 2018. doi: 10.1002/mrm.27472. [3] Prieto C et al, J Magn Reson Imaging. 2015;41(3):738-46. [4] Henningsson M et al, Magn Reson Med. 2012;67(2):437-45. [5] Cruz G et al, Magn Reson Med. 2017 May;77(5):1894-1908.[6] Kellman P et al, Magn Reson Med. 2002 Feb;47(2):372-83.