Authored date:2006-03-01
Over the past decade, thanks to steady advances in pulse sequences and scanner design, contrast-enhanced MR angiography (CE-MRA) has made significant advances in image quality and reliability. For a variety of applications, 3D CE-MRA has advantages over other modalities such as CTA, or digital subtraction angiography (DSA), and in many institutions it is often the imaging test of first choice.
Recently, whole body, 3.0T MR systems have become available, with the promise of greatly improved signal-to-noise ratio (SNR) in comparison to 1.5T. Also, since the longitudinal relaxation time (T1) of background increases with field strength, sensitivity to injected gadolinium agents for CE-MRA is heightened, so that smaller contrast doses may be used. Whilst there are also challenges and disadvantages to 3.0T imaging, such as SAR limitations, dielectric resonances and radiofrequency (RF) eddy currents, nevertheless, CE-MRA at 3.0T can often generate spectacular studies, providing very high spatial resolution 3D data over a large field of view (FoV). This article describes existing 3D CE-MRA techniques at 3.0T, as implemented on the Siemens MAGNETOM Trio scanner. We also summarize clinical experience with CE-MRA to date over a variety of vascular territories.
1. Conventional CE-MRA:
CE-MRA relies on the T1 shortening effect of paramagnetic contrast agents and is performed with a T1-weighted fast spoiled 3D gradient echo (GRE) pulse sequence. The sequence parameters and the contrast administration scheme should be carefully planned, to achieve the best compromise between the expendable signal-to-noise (SNR) and the required spatial and temporal resolution. With gradient slew rates of up to 200 mT/m/s and up to 45 mT/m gradient amplitudes, TRs of the order of 2.5–3 ms and TE of the order of 1.2 ms are achievable, for 512 matrix acquisitions. Pulse sequence features such as asymmetric echo and different k-space sampling schemes have also improved the performance of CE-MRA. Parallel acquisition has greatly increased the performance of CE-MRA at 3.0T, where the higher baseline SNR provides a firmer base to support aggressive acceleration factors. Precise contrast administration is an essential prerequisite for CE-MRA, where the center of k-space should be aligned with the peak vascular enhancement. Timing can be easily optimized by use of a test bolus, but real-time triggering algorithms provide an alternative, if less flexible, method. Contrast doses between 0.1 and 0.2 mmol/kg body weight Gd-based contrast agent have been reported to be sufficient for most single-station MRA examinations.
2. Time-resolved CE-MRA:
Advances in ultrafast MR techniques can now generate temporally resolved 3D MRA, which depict the transit of the paramagnetic contrast agent through the vascular system. Time resolved MRA has the ability to provide supplemental functional information about cardiovascular hemodynamics. Other strengths of time resolved MRA include relative insensitivity to motion and the requirement for only very small doses of contrast.
A general limitation of all time-resolved MRA techniques is that the spatial resolution of the individual 3D data set is limited when compared to single-phase MRA acquisitions. Nonetheless, for many applications, in-plane resolution can be preserved while through-plane resolution is traded for rapid temporal sampling. Unless off-axis reconstruction of the time-resolved data is required (in our experience an uncommon requirement), there is little to be gained from spending a lot of time on through-plane phase-encoding. The MIP images are then viewed in a cine format in the plane of acquisition. Temporal resolution of one second or better is readily achieved with this approach.
The introduction of parallel imaging is one of the most promising recent advances in MR imaging, and it has significantly improved the performance of MRA applications by changing the way that data are acquired and processed. In these techniques, component coil signals in a radiofrequency coil array are used to partially encode spatial information by substituting for phase-encoding gradient steps that have been omitted. Therefore only a subset of the k-space data, defined by the ‘acceleration factor’ is sampled and then the whole dataset is reconstructed afterward. The major drawback to parallel acquisition is that SNR is diminished, and this represents a fundamental challenge as acceleration factors are increased. Among the strategies to counteract the SNR loss of parallel imaging are the use of higher magnetic field, improvement and adjustment in array coil geometry and sensitivity, and the optimal infusion of contrast agent.
The availability of whole body 3.0T MR systems with higher baseline SNR, promises more efficient use of parallel acquisition. Recently integrated multicoil technology has also become available at 3.0T (Tim, Siemens). Appropriately designed array coils with more optimal sensitivity profiles and more channels, will improve overall SNR and CE-MRA performance. 32-channel MR systems are now available at 3.0T, with 102 coil elements which can be sited simultaneously, if required. These multiple RF receiver coils, with associated multi-RF receiver channel electronics, combine for more effective parallel acquisition strategies. The improved CE-MRA performance can be used in any combination to increase coverage, speed or spatial resolution.
Head & Neck
3D CE-MRA in craniocervical vasculature is useful for atherosclerotic arterial disease (Fig.1), aneurysms (Fig.2), AVMs, and pre-operational assessment of tumors.
Fig. 1 Coronal MIP from CE-MRA shows severe stenosis of the left carotid bifurcation and right vertebral artery (arrows) (voxel: 0.8 x 0.7 x 0.7 mm in a 21 second breath hold).
Fig. 2 Coronal MIP and VR from CE-MRA show an arterial aneurysm at the level of the right cavernous ICA (voxel: 0.8 x 0.7 x 0.7 mm in a 21 seconds breath hold).
Fig. 3 MIP images from time resolved CE-MRA show subclavian steal syndrome (left side) (voxel:
1 x 1.2 x 3 mm over a 500 mm FOV, 3D volume was updated every 2 seconds, after injection of 6 ml of contrast).
Pulmonary
Clinical applications for pulmonary CE-MRA include the evaluation of pulmonary embolism, pulmonary perfusion (Fig. 5), pulmonary hypertension (Fig. 6), pulmonary AVM, congenital heart disease, lung tumors, and pulmonary venous mapping (Fig. 7).
Fig. 4 Coronal partial thickness MIP from high spatial resolution pulmonary CE-MRA at 3.0T, acquired during a 20 seconds breath hold, showing up to 5th order pulmonary arterial branches (voxels 0.8 x 0.9 x 1 mm over a 450 mm FOV).
Fig. 5 Coronal MIP images from time resolved MRA at 3.0T demonstrate the sequential filling of the pulmonary and systemic arteries and pulmonary perfusion (voxel size: 1.6 x 1.2 x 5 mm, each 3D volume was updated in 1.5 s).
Fig. 6 Coronal MIP image (A) from CE-MRA at 3.0T shows dilatation of central pulmonary vessels and abnormal proximal-to-distal tapering of the pulmonary arteries. These findings are consistent with pulmonary hypertension which is secondary to a large ventricular septal defect (VSD) (B).
Fig. 7 3D VR image from CE-MRA defines pulmonary venous anatomy, which can improve pre-procedural planning and fluoroscopic procedural time during the ablation treatment for patients with atrial arrhythmias.
Fig. 8 Coronal MIP from CE-MRA at 3.0T showing the irregularity and tortuosity of the abdominal aorta and bilateral renal artery stenosis due to atherosclerosis.
Fig. 9 Sagittal oblique 3D VR from CE-MRA at 3.0T shows aneurysmal dilatation of the superior mesenteric artery (voxel size: 0.8 x 0.8 x 1mm; 21 second breath hold). The origin of the celiac artery is occluded.
Fig. 10 Coronal MIP (A) and 3D VR (B) from CE-MRA at 3.0T demonstrate widely patent transplant renal artery (voxel size: 0.8 x 0.8 x 0.9, FOV: 440, time: 21 s). Time resolved MRA (C) shows an AV-fistula between the right common iliac artery and vein results in early filling of the inferior vena cava. Metal in the right inguinal region from prior surgery causes focal signal loss.