Authored date:2006-02-20

Introduction

Magnetic resonance angiography (MRA) has undergone significant developments over the past decade. For imaging intracranial vessels, time-of-flight (ToF) MRA sequences have been widely used (5, 6, 11). With this technique the vessels give high signal intensity related to the inflow effect of blood during its passage through the acquisition volume, whereas the background tissue appears dark because a short repetition time prevents relaxation of stationary tissue.

Present available pulse sequences are well optimized for ToFMRA at 1.5T. However, one of the principal limitations inherent to ToF-MRA is that they remain signal limited when pushed to the limits of higher resolution and shorter acquisition times. The main advantage of high B-field imaging is a significant improvement in the signal-to-noise-ratio (SNR), which increases in an approximately linear fashion with field strength in the range of 1.5T to 3.0T. Thus, the gain in SNR at 3.0T can be used for either a reduction in imaging time or an increase in resolution. ToF-MRA is a technique that can benefit from the increased SNR available at 3.0T by decreasing voxel size, resulting in improved spatial resolution compared to 1.5T. In addition, advances in coil technology have resulted in further signal gains and multichannel technology has allowed for novel acquisition strategies such as integrated parallel acquisition techniques (iPAT);
(3, 4, 7, 8).

Material and Methods

Seven volunteers and 5 patients (4 aneurysms, 1 AVM) underwent 3D-ToF-MRA at 1.5T (MAGNETOM Sonata) and 3.0T (MAGNETOM Trio) with and without parallel acquisition techniques (iPAT) using similarly designed 8-channel phasedarray head coils. Imaging time of the pulse sequences was set to 7.15 and 7.35 min, respectively. The pulse sequence parameters for both 1.5T and 3.0T are listed in Table 1. Images were analyzed quantitatively by calculating signal-to-noise (SNR) and contrast-to-noise (CNR) ratios of proximal M2 segments and qualitatively by using a 5-point scale.

Tab. 1  SNR and CNR values obtained at 1.5T and 3.0T with and without iPAT. Blood measurements were made on the proximal M2 segments of the MCA.

Results

All patients tolerated the MR examination well. No sensorymotor stimulations or signs related to the increased RF deposition (such as increased sweating) particularly at 3.0T were reported. Analysis of the vessel SNR and CNR is summarized in Table 1. The results indicate a significant increase in both vessel SNR and CNR at 3.0T. ToF-MRA without iPAT at 3.0T showed a higher, statistically insignificant SNR and CNR compared with MRA at 3.0T with iPAT. Both SNR and CNR were significantly higher at 3.0T compared with 1.5T. Overall vessel visualization was rated more highly at both MRA sequences with 3T (Table 2). A slightly higher, statistically not significant overall score was obtained for the MRA with iPAT at 3.0T compared with the 3.0T MRA without iPAT. In particular, visualization of smaller vessel segments, such as M3 and P3 segments as well as delineation of PICA and AICA, was rated significantly superior compared to both MRA sequences at 1.5T (Fig. 1). Wrap around artifacts in the MRAs with iPAT were minor at both 1.5T and 3.0T and had no noticeable influence on image analysis. The increased susceptibility effects at 3.0T, especially at air-bone interfaces along the floor of the anterior cranial fossa and adjacent to the petrous portions of the temporal bones had no noticeable effects on the image quality of ToF MRA at 3.0T. Delineation of a left temporal AVM in one patient was slightly better at both MRAs at 3.0T (Fig. 2). One aneurysm of the right MCA (Fig. 3) with a size of 2.8 mm was reliably detected only at 3.0T, while the other aneurysms – sized between 6 and 10 mm – were detected at both field strengths.

Fig. 1  Axial collapse 3D-ToF MIP images at 3.0T with (A) and without (B) iPAT and at 1.5T with iPAT (C) and without iPAT (D). Note the better visualization of distal MCA and PCA branches as well as the right AICA at 3.0T (A, B). Minor aliasing artifacts are noticed due to the use of the iPAT reconstruction algorithm (A, C).T. 

Fig. 2  One aneurysm of the right MCA with a size of 2.8 mm was reliably detected only at 3.0T.

Fig. 3  Patient with left temporal AVM. Note the superior image quality as well as the superior small vessel delineation at 3.0T (left) compared to 1.5T (right).

Tab. 2  Confidence of vessel visualization at 1.5T and 3.0T with and without iPAT. Five-point scale with scores of 1=unsatisfactory, 2 = fair, 3=average, 4=good, and 5=exellent.

Discussion

ToF-MRA is commonly used in the evaluation of intracranial vascular pathology. Present available pulse sequences are well optimized for ToF-MRA at 1.5T. However, one of the principal limitations inherent to ToF-MRA is that they remain signal limited when pushed to the limits of higher resolution and shorter acquisition times. There are various factors which enhance the overall image quality in ToF-MRA at 3.0T compared with MRA at lower field strengths. The increased SNR available at 3.0T was used in our study to increase spatial resolution, which reduces the amount of partial volume artifact. Smaller voxels are less subject to intravoxel dephasing because they contain a smaller heterogeneity of spins, providing further improvements in MRA. In addition, the effects of magnetic-field strength-related T1-lengthening of brain parenchyma and background tissue are beneficial for ToF-MRA at 3.0T, providing better suppression of background signal. Furthermore, there is little change in the T1 relaxation of blood, making inplane saturation effects similar at 1.5T and 3.0T for ToF techniques (2). The lengthened T1 of background tissue allows ToFMRAat 3.0T with smaller flip angles, decreasing saturation effects within in-plane blood. Since high resolution scanning increases the total number of phase encoding- and 3D-encoding steps, a prolongation of measurement time will result. Owing to the prolonged T1-relaxation constants it was useful to remove the MT pulse in our measurement protocol. MT pulses increase the TR of a measurement protocol by typically 8-12 ms, depending on field strength. Therefore, by removing the MT pulse, a 30% higher resolution within the same total measurement time can be achieved. In addition, the field strength of 3.0T allows an echo time of 3.8 ms to be used as the opposed phase condition. This echo time is shorter than the corresponding 7 ms at 1.5T, which will result in decreased intravoxel dephasing.

In our study, a significantly higher confidence of small vessel visualization was obtained with both ToF-MRA sequences at 3.0T. Despite a lower SNR and CNR, small vessel conspicuity was rated best at 3.0T with iPAT, which was the result of a higher resolution matrix (512 x 640 vs. 296 x 512) and smaller voxel sizes (0.08 vs. 013 mm³) compared with the MRA at 3.0T without iPAT.

Main applications of iPAT are the reduction of examination time by faster imaging or the increase of spatial resolution in a given acquisition time. However, the trade-off for reducing the number of acquired k-space lines using iPAT is a decreased SNR. Owing to this loss of SNR, parallel acquisition techniques are particularly useful when the corresponding image has a high intrinsic SNR like in 3D-ToF-MRA at 3.0T.

Several techniques have been suggested for the iPAT image reconstruction from the reduced data sets (3, 4, 7, 8). They can be divided into two different groups, such as techniques working on the data in frequency domain (e.g. SMASH, GRAPPA) and techniques working on the Fourier transformed data in the image domain (e.g. SENSE). However, one general limitation of the iPAT approach is the propagation of wrap-around artifacts into the center of the image. We therefore decided to use an AUTO-SMASH-like algorithm such as GRAPPA, since these artifacts are less prominent compared to the SENSE technique (4). In our study, aliasing artifacts were only mild and therefore did not limit the accurate assessment of fine vessel detail.

In conclusion, we have demonstrated that 3D-ToF-MRA at 3.0T is superior to that at 1.5T. The combined use of a multichannel phased-array head coil in conjunction with iPAT allows for high resolution intracranial vessel imaging with adequate SNR in reasonable imaging times. With continued optimization and refinements, ToF-MRA at 3.0T will further reduce the need for conventional digital subtraction angiography, which is still an invasive method with possible serious complications.

References

[1] Bernstein MA, Huston J, Lin C, et al. High-resolution intracranial and cervical MRA at 3.0T: technical considerations and initial experience. Magn Reson Med 2001;46:955–62.
[2] Campeau NG, Huston J, Bernstein MA, et al. Magnetic resonance angiography at 3.0 Tesla: initial clinical experience. Topics Magn Reson Imaging 2001;12: 183–204.
[3] Dietrich O, Nikolaou K, Wintersperger BJ, et al. iPAT: applications for fast and cardiovascular MR imaging. Electromedica 2002;70:133–46.
[4] Griswold MA, Jakob PM, Heidemann RM, et al. Generalized autocalibrating partially parallel acquisitions (GRAPPA). Magn Reson Med 2002; 47:1202–10.
[5] Heiserman JE, Drayer BP, Keller PJ, et al. Intracranial vascular stenosis and occlusion: evaluation with threedimensional time-of-flight MR angiography. Radiology 1992;185:667–73.
[6] Marchal G, Bosmans H, van Fraeyenhoven L, et al. Intracranial vascular lesions: optimization and clinical evaluation of three-dimensional time-of-flight MR angiography. Radiology 1990;175:443–8.
[7] Pruessmann KP, Weiger M, Scheidegger MB, et al. SENSE: sensitivity encoding encoding for fast MRI. Magn Reson Med 1999;42:952–62.
[8] Sodickson DK, Manning WJ. Simultaneous acquisition of spatial harmonics (SMASH): fast imaging with radiofrequency coil arrays. Magn Reson Med 1997;38: 591–603.
[9] Thomas SD, Al-Kwifi O, Emery DJ, et al. Application of magnetization transfer at 3.0T in three-dimensional time-of-flight magnetic resonance angiography of the intracranial arteries. J Magn Res Imag 2002;15:479–83.
[10] U.S. Department of Health and Human Services, Food and Drug Administration, Center for Devices and Radiological Health. Guidance for the submission of premarket notifications for magnetic resonance diagnostic devices. Washington, DC: 14 Nov., 1998.
[11] Wentz KU, Roether J, Schwartz A, et al. Intracranial vertebrobasilar system: MR angiography. Radiology 1994;190:
105–10.