Whole-Heart Coronary Magnetic Resonance Angiography at 3 Tesla in 5 Minutes With Slow Infusion of Gd- BOPTA, a High-Relaxivity Clinical Contrast Agent

COMMUNICATION Magnetic Resonance in Medicine 58:1 7 (2007) Whole-Heart Coronary Magnetic Resonance Angiography at 3 Tesla in 5 Minutes With Slow Infusion of Gd- BOPTA, a High-Relaxivity Clinical Contrast Agent Xiaoming Bi, 1,2 James C. Carr, 1 and Debiao Li 1,2 * T 1 -shortening contrast agents have been used to improve the depiction of coronary arteries with breath-hold magnetic resonance angiography (MRA). The spatial resolution and coverage are limited by the duration of the arterial phase of the contrast media passage. In this study we investigated the feasibility of acquiring free-breathing, whole-heart coronary MRA during slow infusion of the contrast media (0.3 ml/s) for prolonged blood signal enhancement time. Ultrashort TR (3 ms) and parallel data acquisition were used to allow the whole-heart MRA in approximately 5 min. A newly approved gadolinium (Gd)- based high T 1 relaxivity contrast agent, gadobenate dimeglumine ([Gd-BOPTA] 2 ), was used and coronary MRA was performed on a whole-body 3 Tesla (T) system to improve the signal-to-noise ratio (SNR). Results from eight volunteers demonstrate that this coronary MRA method is capable of imaging the whole heart in 4.5 0.6 min. Major coronary arteries are well depicted with high SNR (42.4 12.5) and contrast-to-noise ratio (CNR; 27.1 7.6). Magn Reson Med 58:1 7, 2007. 2007 Wiley-Liss, Inc. Key words: magnetic resonance imaging; magnetic resonance angiography; coronary arteries; 3 Tesla; contrast media Paramagnetic T 1 -shortening contrast agents have been used in coronary magnetic resonance angiography (MRA) to improve the signal-to-noise ratio (SNR) and contrast-tonoise ratio (CNR) (1 3). Based on their ability to diffuse to interstitial space, these agents are typically classified as intra- or extravascular agents. Intravascular agents allow for a prolonged half-life in the blood pool, and thus provide long time windows for the acquisition of high-resolution coronary MRA (2 4). However, these agents have not been approved for clinical usage in the United States. All clinically available gadolinium (Gd)-based contrast media are extravascular agents that diffuse to interstitial space quickly. For coronary MRA, such contrast media are typically administered in a short time to ensure adequate T 1 -shortening of the blood pool. The spatial resolution and/or 3D coverage of each contrast-enhanced (CE) scan are limited by confining the data acquisition to a short time window that coincides with the arterial phase of the contrast media passage (1). Images were acquired during a single breath-hold to minimize the imaging time. Gadobenate dimeglumine ([Gd-BOPTA] 2, MultiHance; Bracco Imaging SpA, Milan, Italy) is an extravascular contrast agent that was recently approved for clinical use in the United States and Europe. Because of this agent s capacity for weak and transient interaction with serum albumin, the disposition of the Gd-BOPTA molecule leads to an elevated vivo T 1 relaxation rate (R 1 9.7 (mmol/l) 1 s 1 at 1.5 Tesla (T)) compared to that of the conventional Gd-based clinical agents (between 4.3 and 5.6 (mmol/l) 1 s 1 ) (5). Such relaxivity is only slightly reduced from 1.5T to 3T, as reported in a previous paper (6). Thus, it is possible to achieve T 1 shortening similar to that obtained with conventional Gd agents at a lower concentration. 3T has been shown to be a promising platform for performing coronary MRA, due to the SNR gain over 1.5T (7 9). However, this SNR gain has not been directly translated into improved coronary artery delineation, mainly due to increased image artifacts at 3T with steady-state free precession (SSFP) imaging. Conventional spoiled gradient-echo sequences, such as fast low-angle shot (FLASH), are relatively insensitive to the increased field inhomogeneities at 3T. The combination of such a T 1 -weighted sequence and high-relaxivity contrast agent could potentially be a powerful method for performing coronary MRA at 3T. In this work we propose a new methodology for acquiring whole-heart MR images of the coronary arteries in approximately 5 min. High-resolution images were acquired at 3T with slow infusion of the clinical contrast agent Gd-BOPTA. A magnetization-prepared, navigatorand electrocardiogram (ECG)-gated FLASH sequence was employed for data acquisition during free breathing. The feasibility of this method was validated in healthy volunteers. 1 Department of Radiology, Northwestern University, Chicago, Illinois, USA. 2 MATERIALS AND METHODS Department of Biomedical Engineering, Northwestern University, Chicago, Illinois, USA. Eight healthy adults (four males and four females, 32.8 Grant sponsor: National Institutes of Health; Grant number: NIBIB EB002623; 8.7 years old) were imaged on a 3T whole-body scanner Grant sponsors: Siemens Medical Solutions USA, Inc.; Bracco Diagnostics, Inc.; Ritcher Fellowship. (Trio; Siemens Medical Solutions, Erlangen, Germany). *Correspondence to: Debiao Li, Ph.D., Suite 700, 448 East Ontario St., Chicago, IL 60611. E-mail: d-li2@northwestern.edu rate of 200 mt/m/ms and a maximum gradient strength of The system was capable of operating at a maximum slew Received 1 August 2006; revised 14 December 2006; accepted 5 February 40 mt/m. Twelve-element matrix coils (six anterior and 2007. DOI 10.1002/mrm.21224 six posterior) equipped with the scanner were activated for Published online in Wiley InterScience (www.interscience.wiley.com). data acquisition. 2007 Wiley-Liss, Inc. 1

2 Bi et al. Precontrast Imaging All imaging acquisitions were performed under free breathing with the subjects in a supine position. Twodimensional (2D) scout images were first obtained in three orthogonal orientations using a low-resolution, single-shot FLASH sequence (repetition time (TR) 3.2 ms, echo time (TE) 1.5 ms, matrix 192 192, field of view (FOV) 400 400 mm, slice thickness 6 mm, flip angle (FA) 12 ). These scout images were used to identify the position of the heart and diaphragm. A cine scan was then prescribed in the four-chamber view to determine the quiescent period for coronary artery imaging (10). Parameters for the cine scout included TR/TE 5/2.4 ms, FOV 255 340 mm, matrix 111 192, slice thickness 6 mm, and FA 15. Images were acquired with a generalized autocalibrating partially parallel acquisition (GRAPPA) (11) acceleration factor of 2. Retrospective ECG gating was applied and four image acquisitions were averaged to minimize respiratory motion-related artifacts. Fifty cardiac phases were reconstructed for visual assessment of global cardiac motion to determine the trigger delay time and the duration of acquisition window per heartbeat. A navigator-gated, ECG-triggered, fat-saturated, segmented FLASH sequence was employed for whole-heart coronary artery imaging. Navigator pulses were placed on the dome of the right hemidiaphragm with a 6-mm acceptance window. Prospective real-time adaptive motion correction was applied with a correction factor of 0.6 in the superior inferior direction (12). The 3D k-space data were collected with a centric ordering scheme in the phaseencoding direction, and linear order in the partition-encoding direction. Eighty transverse slices were acquired and interpolated to 160 slices of 0.75 mm thickness. The in-plane spatial resolution was 0.65 0.70 mm 2 (interpolated from 1.3 1.4 mm 2 ). To speed up the image acquisition, parallel data acquisition (GRAPPA) was used in the phase-encoding direction with an acceleration factor of 2. Other imaging parameters included TR/TE 3.0/1.4 mm, flip angle 15, lines per heartbeat 47, and readout bandwidth 610 Hz/pixel. The matrix and FOV were 256 141 80 (interpolated to 512 282 160) and 330 200 120, respectively (readout phase-encoding partition-encoding direction). Additional images were acquired in three volunteers with 40-ms T 2 preparation after a B 1 -insensitive T 2 preparation (13) was implemented on the scanner. CE Imaging Coronary CE-MRA was performed using the same sequence and similar parameters as precontrast imaging. A nonselective inversion pulse was applied prior to the navigator-echo pulses to suppress the background signal. Magnetization was selectively reinverted after the inversion pulse to restore the magnetization for navigator-echo pulses. Gd-BOPTA (0.2 mmol/kg body weight) was slowly injected from the antecubital fossa using a power injector (Spectris; Medrad, Indianola, PA, USA) at a rate of 0.3 ml/s. Infusion of contrast agent was immediately followed by 20 ml of saline solution injected at the same rate. Data acquisition was started 25 s after the initialization of contrast agent administration. The purpose of applying an inversion pulse was to suppress the signal from myocardium. With the slow infusion of contrast agent, it was expected that the myocardium T 1 value would be slightly decreased over time. Our initial tests showed that an inversion time (TI) value of 200 ms yielded effective myocardium signal suppression and maintained high blood signal intensity (SI). This TI value was used in CE imaging of all subjects. Image Reformatting and Analysis Collected 3D coronary artery images were reformatted using image-processing software InSpace (Siemens Medical Solutions, Erlangen, Germany). Curved multiplanar reconstruction (MPR) was performed along the coronary artery course. In addition, whole-heart coronary MR angiograms were reformatted using the CoronaViz software (Siemens Corporate Research, Princeton, NJ, USA) to project multibranch vessels with their surroundings onto a single image (14). The blood SI, myocardial SI, and background noise were measured from original images for SNR and CNR calculations. A circular region of interest (ROI; area 100 mm 2 ) for measuring the blood SI was placed in the aorta at the level of the left coronary ostium. Myocardial SI was measured from the connective tissue immediately next to the coronary arteries (ROI area 60 80 mm 2 ). Image noise was estimated to be the standard deviation (SD) of a circular ROI (area 180 mm 2 ) placed in the background air (9). The SNRs of the blood and myocardium were calculated by dividing the mean SI of the tissue region by the image noise. The CNR was calculated by subtracting the SNR of the myocardium from that of the blood. In addition, the lengths of depicted coronary arteries were measured using previously described methods (1). Images acquired without and with contrast agent were blindly graded by two experienced reviewers. Reformatted images of the left and right coronary arteries were placed in a random order and scored on a system of 1 4 (15): 1 poor or uninterpretable (coronary artery with markedly blurred borders or edges), 2 good (coronary artery visible and moderately blurred), 3 very good (coronary artery visible and mildly blurred), and 4 excellent (coronary artery visible and sharply defined). The quantitative results of the measurements are presented as the mean SD. Comparisons of the SNR, CNR, depicted vessel lengths, and image quality between preand postcontrast scans were performed using a paired two-sample t-test, with a two-tailed P-value 0.05 considered statistically significant. RESULTS Volunteer studies were successfully performed without complications. The major coronary arteries were depicted in all subjects with administration of contrast media. Figure 1 illustrates a set of exemplary coronary images acquired from a 40-year-old, 192-pound volunteer. The depiction of the right coronary artery (RCA) and left circum-

Whole-Heart Coronary CE-MRA at 3T 3 FIG. 1. Exemplary coronary artery images acquired from a 40-year-old, 192-pound volunteer without (a) and with (b) slow infusion of contrast agent. Note that the depiction of the RCA is markedly improved with administration of contrast agent. The proximal and middle sections of the LCX are clearly visualized as well in the postcontrast image. flex (LCX) artery is markedly improved with the administration of Gd-BOPTA. Figure 2 shows a pair of axial images extracted from preand postcontrast whole-heart 3D image sets of the same volunteer. The blood myocardial contrast is substantially improved with slow infusion of Gd-BOPTA, as indicated by signal profiles across the blood myocardial boundary. Note that the cross sections of the left anterior descending (LAD) artery and RCA are sharply depicted in the CE image. In addition, apparent respiratory artifacts in the precontrast image were dramatically suppressed by applying the nonselective inversion pulse in the postcontrast scan. Reformatted RCA images of a subject obtained with T 2 - prepared FLASH (no contrast agent given) and slow infusion of contrast agent are illustrated in Fig. 3. Compared to the precontrast image acquired with T 2 preparation, the RCA is sharply depicted with the administration of contrast media. The average infusion time for contrast agent and saline is 170 s. The measured results of the SNR, CNR, depicted vessel lengths, and imaging time are summarized in Table 1. With similar imaging times for pre- and postcontrast scans, the SNR, CNR, and depicted vessel length are significantly improved with slow infusion of Gd-BOPTA. The average imaging time for whole- FIG. 2. Axial images extracted from whole-heart data sets without (a) and with (b) the administration of Gd-BOPTA. Image c illustrates the profile of SI of a cross line (line width 10 pixels) running through the blood and myocardium as illustrated in b. The blood myocardial contrast is substantially improved with slow infusion of Gd-BOPTA, as shown in b and c. The cross sections of the RCA and LAD are well depicted (solid arrows in b). Also note that apparent respiratory artifacts in the precontrast image (dashed arrows in a) are eliminated by the nonselective inversion pulse in the postcontrast scan.

4 Bi et al. FIG. 3. Reformatted RCA images from two whole-heart measurements with T 2 -prepared FLASH (a) (no contrast agent given) and inversion-prepared FLASH (b) with slow infusion of contrast agent. Note the markedly improved depiction of the coronary artery with contrast-agent administration. heart coronary CE-MRA was 4.5 0.6 min in the eight volunteers, with 53% average navigator-gating efficiency. The image quality is substantially improved with slow infusion of the contrast agent. As illustrated in Fig. 4, the average image score (left coronary artery: 1.72 0.82/ 3.25 0.53; right coronary artery: 1.47 0.47/2.97 0.72 for pre-/postcontrast images, respectively) is significantly higher with administration of contrast media. Examples of reformatted whole-heart coronary artery images obtained using volume rendering and CoronaViz software are illustrated in Figs. 5 and 6, respectively. The major coronary arteries are clearly depicted in both figures. The distal segments and some small branches are also visible in the reformatted images. DISCUSSION This study demonstrates the feasibility of performing whole-heart coronary MRA at 3T in 5 min. A decreased imaging time (compared to a conventional protocol with an imaging time of more than 10 min (16)) is useful for reducing potential image artifacts due to inconsistent cardiac and respiratory motion. FLASH data acquisition allowed an ultrashort TR to be used to speed up data acquisition as compared to SSFP. The SNR penalty due to the large bandwidth was compensated for by CE imaging. The prolonged contrast-enhancement time was made possible by slow infusion of Gd-BOPTA, a newly approved clinical contrast agent with high T 1 relaxivity. SSFP has been the method of choice for coronary MRA at 1.5T because it provides SNR and CNR gains over spoiled gradient-echo sequences. Technical challenges remain, however, to apply the same technique at 3T, because of increased field inhomogeneities, energy deposition, distortions of radiofrequency (RF) pulses, and gradients from dielectric resonance and eddy currents. Various methods have been demonstrated to be useful for alleviating SSFP sequence-related imaging artifacts, such as shifting the synthesizer frequency to reduce off-resonance-related image artifacts (17), or improving field homogeneity by applying localized linear or second-order shimming (18). Other studies showed improved uniformity of T 2 preparation using adiabatic refocusing pulses (13), preconditioning the spin system with a starter sequence (19), or reducing flow- and eddy-currentinduced phase errors by reordering data acquisition (20). These measures have improved SSFP coronary MRA at 3T to a certain degree; nevertheless, certain limits and tradeoffs exist. For instance, improved T 2 preparation and fat saturation using adiabatic pulses increase power deposition and may limit the imaging FAs. The effectiveness of frequency shift and local shimming vary among individual subjects as Table 1 Quantitative Results of Coronary MRA From 8 Volunteers Depicted vessel length (cm) Imaging sequence SNR CNR Imaging time (min) LAD LCX RCA FLASH 21.7 13.3 1.4 0.8 4.8 1.5 7.8 2.5 3.1 0.9 8.1 2.8 T 2 prep FLASH (n 3) 19.9 5.4 4.9 2.1 5.2 2.1 8.1 0.9 3.5 0.8 9.1 1.3 IR-FLASH with contrast 42.4 12.5* 27.1 7.6* 4.5 0.6 10.2 1.7* 6.1 1.2 13.0 4.2* Note that SNR, CNR, and depicted vessel length are significantly improved with slow infusion of Gd-BOPTA as compared to precontrast FLASH. *P 0.05.

Whole-Heart Coronary CE-MRA at 3T 5 FIG. 4. Image quality of the left and right coronary arteries as scored from the images of eight volunteers. With slow infusion of contrast agent, the depiction of the left and right coronary arteries is significantly improved. well. Further technical improvements are required to improve the consistency of image quality of SSFP coronary MRA at 3T. Compared to SSFP, FLASH is relatively insensitive to B 0 field inhomogeneities due to its spoiled gradient structure. Both myocardium and epicardial fat signals can be effectively suppressed by a nonselective inversion pulse, which is less sensitive to B 0 and B 1 inhomogeneities than T 2 preparation and the spectrally selective fat saturation conventionally used with SSFP. Therefore, CE-FLASH is likely to be more tolerant of imperfections in B 0 and B 1 fields, and result in more consistent image quality among subjects at 3T. In addition, CE-FLASH deposits less power than SSFP because it uses lower FAs and eliminates T 2 preparation, which requires a composite RF pulse train or high-energy adiabatic refocusing pulses. It is known that SNR measurement with parallel data acquisition can be affected by the geometry (g)-factor of the coil and treatment of data during parallel image reconstruction. Since the coil configurations, targeted imaging volume, and image reconstruction method were identical for both pre- and postcontrast measurements, we expect the g-factor of the coils to have similar effects on pre- and postcontrast SNRs measured in the same ROIs. Therefore, the SNR and CNR calculated from pre- and postcontrast images were compared in relative terms to assess the effect of the contrast agent. In addition, image quality was compared between pre- and postcontrast imaging. T 2 preparation has been widely used in coronary MRA at 1.5T to improve the blood myocardium contrast. At 3T, however, a conventional Malcom-Levit (MLEV)-weighted T 2 preparation scheme (21) resulted in substantial image artifacts in our initial tests due to increased B 0 and B 1 inhomogeneities. Therefore, we started the study without T 2 preparation for precontrast imaging. The utilization of adiabatic pulses for refocusing has been shown to improve T 2 preparation at 3T (13). This preparation scheme was implemented in the middle of this study and applied to three volunteers. With T 2 preparation, the SNR was reduced and CNR increased, as shown in Table 1 and Fig. 3. The postcontrast SNR, CNR, and increase of these values from precontrast to postcontrast observed in this study are somewhat higher than those obtained in most previous studies at 1.5T. First, 3T imaging has higher SNR and CNR than 1.5T imaging (7). Second, the MultiHance used in the present study has higher T 1 relaxivity than conventional Gd contrast agents. Third, the T 1 value is increased at 3T as compared to 1.5T, which allows more complete background suppression. Finally, an ultrashort TR of 3 ms was used to speed up data acquisition, which limits the pre- FIG. 5. a and b: Volume-rendered images from a 4:14-min postcontrast whole-heart MRA scan. The RCA and LAD are well depicted. The corresponding curved MPR image is shown in image c, from which the depicted vessel length is measured.

6 Bi et al. FIG. 6. Whole-heart coronary artery images reformatted using CoronaViz software. Note that the left and right coronary arteries are sharply depicted and the distal segments and small branches are visible, as indicated by arrows. contrast SNR and allows greater SNR increases in postcontrast imaging. The scan time used in this study was shorter than that employed in most similar studies at 1.5T (in the range of 10 15 min to cover the whole heart). Recent studies using 32-channel coils (22,23) have shown great potential for reducing the imaging time for coronary MRA with acceleration factors up to 16. In the current experiment, an acceleration factor of 2 was used to achieve whole-heart coverage in 4.5 min. This time could potentially be reduced with further acceleration (e.g., 2D parallel imaging, partial Fourier, or higher acceleration factor) and an increased number of coil elements. With reduced imaging time, the injection rate of contrast agent could potentially be increased to compensate for some of the signal loss resulting from increased undersampling. Gd-BOPTA has a roughly twofold higher T 1 relaxivity in blood compared to other clinical Gd-based contrast agents currently in widespread use. A previous intraindividual crossover comparison showed that Gd-BOPTA at a dose of 0.1 mmol/kg is comparable to gadopentetate dimeglumine at a dose of 0.2 mmol/kg for CE imaging of the renal arteries (24). Another study (25) concluded that weak protein binding of Gd-BOPTA can substantially increase the efficacy of such Gd chelates. In that study the enhancement of blood SI 5 min after Gd-BOPTA administration was 182% relative to the same dose of gadopentetate. The higher in vivo relaxivity and prolonged half-life of Gd- BOPTA make it more suitable for such slow-infusion studies compared to conventional Gd-based agents. A previous study (26) showed that intravenous injection of Gd-BOPTA at a high dose (0.2 mmol/kg body weight) and low rate (0.5 ml/s) leads to a long plateau of signal enhancement, which is desirable for coronary CE-MRA. In our study the infusion rate was reduced to 0.3 ml/s; therefore, the contrast-injection duration was approximately half of the data acquisition time and peak blood signal occurred during central k-space data acquisition in the partition-encoding direction, which is in the middle of the imaging time with linear encoding. A quantitative assessment of the enhancement kinetics of contrast media is necessary to optimize the data-acquisition strategy and contrast-injection scheme. In conclusion, whole-heart coronary MRA in 5 min at 3T with slow infusion of a clinical contrast agent is feasible. Further improvement in imaging speed with parallel acquisition may allow an increased contrast-infusion rate and higher SNR. ACKNOWLEDGMENTS The authors thank Sharon Coffey for assistance in contrast administration. X.B. acknowledges the support of the Ritcher Fellowship. REFERENCES 1. Li D, Carr JC, Shea SM, Zheng J, Deshpande VS, Wielopolski PA, Finn JP. Coronary arteries: magnetization-prepared contrast-enhanced threedimensional volume-targeted breath-hold MR angiography. Radiology 2001;219:270 277. 2. Paetsch I, Jahnke C, Barkhausen J, Spuentrup E, Cavagna F, Schnackenburg B, Huber M, Stuber M, Fleck E, Nagel E. Detection of coronary stenoses with contrast enhanced, three-dimensional free breathing coronary MR angiography using the gadolinium-based intravascular contrast agent gadocoletic acid (B-22956). J Cardiovasc Magn Reson 2006; 8:509 516. 3. Stuber M, Botnar RM, Danias PG, McConnell MV, Kissinger KV, Yucel EK, Manning WJ. Contrast agent-enhanced, free-breathing, three-dimensional coronary magnetic resonance angiography. J Magn Reson Imaging 1999;10:790 799. 4. Huber ME, Paetsch I, Schnackenburg B, Bornstedt A, Nagel E, Fleck E, Boesiger P, Maggioni F, Cavagna FM, Stuber M. Performance of a new gadolinium-based intravascular contrast agent in free-breathing inversion-recovery 3D coronary MRA. Magn Reson Med 2003;49:115 121. 5. Knopp MV, Runge VM, Essig M, Hartman M, Jansen O, Kirchin MA, Moeller A, Seeberg AH, Lodemann KP. Primary and secondary brain tumors at MR imaging: bicentric intraindividual crossover comparison of gadobenate dimeglumine and gadopentetate dimeglumine. Radiology 2004;230:55 64. 6. Rohrer M, Bauer H, Mintorovitch J, Requardt M, Weinmann HJ. Comparison of magnetic properties of MRI contrast media solutions at different magnetic field strengths. Invest Radiol 2005;40:715 724. 7. Bi X, Deshpande V, Simonetti O, Laub G, Li D. Three-dimensional breathhold SSFP coronary MRA: a comparison between 1.5T and 3.0T. J Magn Reson Imaging 2005;22:206 212. 8. Sommer T, Hackenbroch M, Hofer U, Schmiedel A, Willinek WA, Flacke S, Gieseke J, Traber F, Fimmers R, Litt H, Schild H. Coronary MR angiography at 3.0 T versus that at 1.5 T: initial results in patients suspected of having coronary artery disease. Radiology 2005;234:718 725. 9. Stuber M, Botnar RM, Fischer SE, Lamerichs R, Smink J, Harvey P, Manning WJ. Preliminary report on in vivo coronary MRA at 3 Tesla in humans. Magn Reson Med 2002;48:425 429. 10. Weber OM, Martin AJ, Higgins CB. Whole-heart steady-state free precession coronary artery magnetic resonance angiography. Magn Reson Med 2003;50:1223 1228. 11. Griswold MA, Jakob PM, Heidemann RM, Nittka M, Jellus V, Wang J, Kiefer B, Haase A. Generalized autocalibrating partially parallel acquisitions (GRAPPA). Magn Reson Med 2002;47:1202 1210. 12. Wang Y, Ehman RL. Retrospective adaptive motion correction for navigator-gated 3D coronary MR angiography. J Magn Reson Imaging 2000; 11:208 214. 13. Nezafat R, Stuber M, Ouwerkerk R, Gharib AM, Desai MY, Pettigrew RI. B1-insensitive T2 preparation for improved coronary magnetic resonance angiography at 3 T. Magn Reson Med 2006;55:858 864.

Whole-Heart Coronary CE-MRA at 3T 7 14. Aharon S, Oksuz O, Lorenz C. Simultaneous projection of multibranched vessels with their surroundings on a single image from coronary MRA. In: Proceedings of the 14th Annual Meeting of ISMRM, Seattle, WA, USA, 2006 (Abstract 365). 15. Kim WY, Danias PG, Stuber M, Flamm SD, Plein S, Nagel E, Langerak SE, Weber OM, Pedersen EM, Schmidt M, Botnar RM, Manning WJ. Coronary magnetic resonance angiography for the detection of coronary stenoses. N Engl J Med 2001;345:1863 1869. 16. Sakuma H, Ichikawa Y, Suzawa N, Hirano T, Makino K, Koyama N, Van Cauteren M, Takeda K. Assessment of coronary arteries with total study time of less than 30 minutes by using whole-heart coronary MR angiography. Radiology 2005;237:316 321. 17. Deshpande VS, Shea SM, Li D. Artifact reduction in true-fisp imaging of the coronary arteries by adjusting imaging frequency. Magn Reson Med 2003;49:803 809. 18. Schar M, Kozerke S, Fischer SE, Boesiger P. Cardiac SSFP imaging at 3 Tesla. Magn Reson Med 2004;51:799 806. 19. Foxall DL. Starter sequence for steady-state free precession imaging. Magn Reson Med 2005;53:919 929. 20. Bi X, Park J, Deshpande V, Li D. Reduction of eddy currents induced image artifacts in coronary MRA using a linear-centric-encoding (LCE) SSFP sequence. In: Proceedings of the 13th Annual Meeting of ISMRM, Miami, FL, USA, 2005 (Abstract 1612). 21. Brittain JH, Hu BS, Wright GA, Meyer CH, Macovski A, Nishimura DG. Coronary angiography with magnetization-prepared T2 contrast. Magn Reson Med 1995;33:689 696. 22. Nehrke K, Bornert P, Mazurkewitz P, Winkelmann R, Grasslin I. Freebreathing whole-heart coronary MR angiography on a clinical scanner in four minutes. J Magn Reson Imaging 2006;23:752 756. 23. Niendorf T, Hardy CJ, Giaquinto RO, Gross P, Cline HE, Zhu Y, Kenwood G, Cohen S, Grant AK, Joshi S, Rofsky NM, Sodickson DK. Toward single breath-hold whole-heart coverage coronary MRA using highly accelerated parallel imaging with a 32-channel MR system. Magn Reson Med 2006;56:167 176. 24. Prokop M, Schneider G, Vanzulli A, Goyen M, Ruehm SG, Douek P, Dapra M, Pirovano G, Kirchin MA, Spinazzi A. Contrast-enhanced MR angiography of the renal arteries: blinded multicenter crossover comparison of gadobenate dimeglumine and gadopentetate dimeglumine. Radiology 2005;234:399 408. 25. Cavagna FM, Maggioni F, Castelli PM, Dapra M, Imperatori LG, Lorusso V, Jenkins BG. Gadolinium chelates with weak binding to serum proteins. A new class of high-efficiency, general purpose contrast agents for magnetic resonance imaging. Invest Radiol 1997;32:780 796. 26. Knopp MV, Schoenberg SO, Rehm C, Floemer F, von Tengg-Kobligk H, Bock M, Hentrich HR. Assessment of gadobenate dimeglumine for magnetic resonance angiography: phase I studies. Invest Radiol 2002; 37:706 715.