(Radiology. 2000;217:581-586.)
© RSNA, 2000
Subclavian MR Arteriography: Reduction of Susceptibility Artifact with Short Echo Time and Dilute Gadopentetate Dimeglumine1
Mohammed A. Neimatallah, MD,
Thomas L. Chenevert, PhD,
Ruth C. Carlos, MD,
Frank J. Londy, RT,
Qian Dong, MD,
Martin R. Prince, MD, PhD and
H. Myra Kim, ScD
1 From the Departments of Radiology (M.A.N., T.L.C., R.C.C., F.J.L.) and Biostatistics (H.M.K.), University of Michigan, Ann Arbor, and the Department of Radiology, Weill Medical College of Cornell University, Box 141, Cornell MRI, 416 E 55th St, New York, NY 10022 (Q.D., M.R.P.). Received June 7, 1999; revision requested July 29; revision received December 29; accepted January 12, 2000. Supported in part by grants from the Whitaker Foundation and the Robert Wood Johnson Clinical Scholars Program. Address correspondence to M.R.P. (e-mail: map2008@mail.med.cornell.edu).
 |
ABSTRACT
|
|---|
At arterial phase gadolinium-enhanced magnetic resonance (MR) angiography, artifactual stenosis of the subclavian artery is sometimes seen adjacent to the subclavian vein on the side of the contrast material injection. Experiments in phantoms and in 19 patients showed increased artifact with longer echo time and higher concentration of injected contrast material. An effective method to substantially decrease this susceptibility artifact was threefold dilution of gadopentetate dimeglumine and use of a short echo time (1 msec).
Index terms: Arteries, MR, 562.12142 • Arteries, subclavian, 562.12142 • Magnetic resonance (MR), artifact • Magnetic resonance (MR), vascular studies, 562.12142 • Veins, MR, 562.12142 • Veins, subclavian, 562.12142
|
INTRODUCTION
|
|---|
Magnetic resonance (MR) angiography is increasingly performed with dynamic contrast material enhancement (1–12). Compared with conventional time-of-flight and phase-contrast MR techniques, this method has the advantages of producing MR images with higher spatial resolution and higher signal-to-noise ratio in substantially less time. Because imaging is dependent on the T1 shortening effect of gadolinium rather than on inflow phenomena, there are fewer flow-related artifacts.
One artifact that is unique to dynamic contrast enhancement is caused by the highly concentrated gadolinium in the vein being injected. In MR angiography of the chest, artifactual signal intensity loss is seen in the subclavian artery adjacent to the subclavian vein ipsilateral to the site of intravenous injection (Fig 1). This signal dropout in the artery can give the false impression of subclavian arterial stenosis or occlusion. We hypothesize that this artifact is due to a susceptibility effect from the highly concentrated gadolinium in the adjacent vein. The strong magnetic field gradient produced by the concentrated gadolinium may induce spin dephasing substantially beyond the subclavian venous lumen so the adjacent artery is affected. If this hypothesis is true, then the artifact should be reduced by diluting the gadolinium and using a shorter echo time.
View larger version (121K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1a. Maximum intensity projection images from coronal, three-dimensional, spoiled gradient-echo, gadolinium-enhanced MR angiograms (5/1, 45° flip angle) depict the subclavian arteries in two patients with a left arm injection. (a, c) Arterial phase images depict signal dropout at the left subclavian artery (right arrow in a, arrow in c) owing to susceptibility artifact from concentrated gadopentetate dimeglumine in the left subclavian vein ipsilateral to the injection site. (a) Signal dropout is seen at the brachiocephalic artery (left arrow) from the concentrated gadopentetate dimeglumine in the left brachiocephalic vein. (b, d) The artifact disappears in the subsequent equilibrium phase.
|
|
View larger version (129K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1b. Maximum intensity projection images from coronal, three-dimensional, spoiled gradient-echo, gadolinium-enhanced MR angiograms (5/1, 45° flip angle) depict the subclavian arteries in two patients with a left arm injection. (a, c) Arterial phase images depict signal dropout at the left subclavian artery (right arrow in a, arrow in c) owing to susceptibility artifact from concentrated gadopentetate dimeglumine in the left subclavian vein ipsilateral to the injection site. (a) Signal dropout is seen at the brachiocephalic artery (left arrow) from the concentrated gadopentetate dimeglumine in the left brachiocephalic vein. (b, d) The artifact disappears in the subsequent equilibrium phase.
|
|
View larger version (155K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1c. Maximum intensity projection images from coronal, three-dimensional, spoiled gradient-echo, gadolinium-enhanced MR angiograms (5/1, 45° flip angle) depict the subclavian arteries in two patients with a left arm injection. (a, c) Arterial phase images depict signal dropout at the left subclavian artery (right arrow in a, arrow in c) owing to susceptibility artifact from concentrated gadopentetate dimeglumine in the left subclavian vein ipsilateral to the injection site. (a) Signal dropout is seen at the brachiocephalic artery (left arrow) from the concentrated gadopentetate dimeglumine in the left brachiocephalic vein. (b, d) The artifact disappears in the subsequent equilibrium phase.
|
|
View larger version (139K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1d. Maximum intensity projection images from coronal, three-dimensional, spoiled gradient-echo, gadolinium-enhanced MR angiograms (5/1, 45° flip angle) depict the subclavian arteries in two patients with a left arm injection. (a, c) Arterial phase images depict signal dropout at the left subclavian artery (right arrow in a, arrow in c) owing to susceptibility artifact from concentrated gadopentetate dimeglumine in the left subclavian vein ipsilateral to the injection site. (a) Signal dropout is seen at the brachiocephalic artery (left arrow) from the concentrated gadopentetate dimeglumine in the left brachiocephalic vein. (b, d) The artifact disappears in the subsequent equilibrium phase.
|
|
In this study, a phantom was used to explore how this artifact varied with gadolinium concentration and echo time. On the basis of phantom data, optimized gadolinium dilution and imaging parameters were studied in 19 patients.
|
Materials and Methods
|
|---|
Phantom
Eleven glass test tubes (8-mm diameter) containing dilutions of gadopentetate dimeglumine (Magnevist [0.5 mol/L]; Berlex Laboratories, Wayne, NJ) with concentrations of 0% (distilled water), 0.1%, 1%, 5%, 10%, 20%, 30%, 40%, 50%, 60%, and 100% (nondilute gadopentetate dimeglumine) were embedded vertically in an agar phantom doped with of 1 mmol/L gadopentetate dimeglumine (Fig 2). The phantom was imaged with a 1.5-T magnet (Horizon; GE Medical Systems, Milwaukee, Wis) with a body coil. To obtain the shortest possible echo time, the bandwidth was maximized. The widest bandwidth on our magnet was 62.5 kHz. In addition, thick partitions (3 mm), large field of view (36–42 cm), and fractional echoes were used to obtain the shortest possible echo time.
View larger version (126K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2. Top left: Schematic of the phantom used in this study illustrates the relative position of the glass tubes with different concentrations of gadopentetate dimeglumine (0%, 0.1%, 1%, 5%, 10%, 20%, 30%, 40%, 50%, 60%, 100%). Subsequent three-dimensional, spoiled gradient-echo MR images (15/1-10, 45° flip angle, 62.5-kHz bandwidth) depict increasing artifact as the echo time (TE) and concentration increase. Notice increased signal intensity within tubes at lower concentrations (30% or less at echo time of 1 msec [TE=1] and 5% or less at echo time of 10 msec [TE=10]).
|
|
MR imaging was performed with a three-dimensional spoiled gradient-echo sequence with the following parameters: fixed repetition time of 15 msec and variable echo times of 1, 2, 3, 5, or 10 msec (15/1, 2, 3, 5, 10), flip angle of 45°, bandwidth of 62.5 kHz, matrix of 256 x 160, field of view of 34 cm, 36 partitions, and partition thickness of 3.3 mm. The artifact around the tube and the tube signal intensity were measured in regions of interest manually defined by one radiologist (M.A.N.) by using a computer workstation (Advantage Windows; GE Medical Systems).
Patients
Nineteen patients (Six male patients and 13 female patients; age range, 1–74 years; mean age, 49.8 years) referred to undergo MR angiography in the thoracic aorta or subclavian artery from September through November 1998 were invited to participate in this study. The study was approved by the institutional review board, and written informed consent was obtained from each patient.
MR angiography was performed by injecting one of three concentrations of gadopentetate dimeglumine (25%, 33%, and 100% [nondilute gadopentetate dimeglumine]). Normal saline solution was used for the dilutions. On the basis of results in the phantom study, concentrations of 25% and 33% were predicted to substantially reduce the artifact but still allow rapid injection of the entire dose of gadopentetate dimeglumine with an injection rate of 3–4 mL/sec. Three patients received 25% concentration; eight patients, 33% concentration; and eight patients, 100% concentration (nondilute gadopentetate dimeglumine). All injections and MR examinations were performed by the same radiologist (M.A.N.).
First, the patients underwent sagittal MR imaging with a half-Fourier rapid acquisition with relaxation enhancement, or RARE, sequence (single-shot fast spin echo). Then, coronal, three-dimensional, gadolinium-enhanced MR angiography was performed with a three-dimensional spoiled gradient-echo sequence with the following parameters: 5.7–7.0/1.0–2.1, section thickness of 2.8–3.4 mm, flip angle of 45°, bandwidth of 62.5 kHz, matrix of 265 x 128–160, and field of view of 33–48 cm. These parameters were adjusted to allow imaging to cover the thoracic aorta and the subclavian arteries within the patient’s breath-hold capacity. Thus, acquisition time for each phase ranged from 27 to 40 seconds.
The imaging delay between initiation of gadopentetate dimeglumine infusion and initiation of three-dimensional imaging was determined by using an automated bolus detection pulse sequence (13) (MR Smartprep; GE Medical Systems). With use of a 10 cm x 40 mm volume, the tracker (region of interest) for automatic bolus detection was positioned over the descending thoracic aorta just above the level of the diaphragm. Centric phase encoding was used to ensure synchronization of the arterial phase of the bolus with central k-space data acquisition (14,15). Use of centric phase encoding also ensured that central k-space data were acquired at the beginning of the breath hold to minimize motion artifact in the event that the patient could not suspend breathing for the entire study.
This sequence was repeated three times: before injection of gadopentetate dimeglumine and after injection, during the arterial and equilibrium phases. The patient was instructed to take several deep breaths between the arterial and equilibrium phases, which resulted in approximately 10–20-second delays between these data acquisitions. A large caliber (18-gauge) angiocatheter was used to allow an injection rate of 3 mL/sec for the 60 mL of 33% dilute gadopentetate dimeglumine, 4 mL/sec for the 80 mL of 25% dilute gadopentetate dimeglumine, and 1 mL/sec for the 20 mL of nondilute gadopentetate dimeglumine. When possible, intravenous access was obtained in the right arm because it has the most direct path to the right atrium while avoiding the left brachiocephalic vein, which may be pinched between an ectatic aorta and the sternum (16). To ensure there was no gap in the bolus, we used a specialized tubing set (Smartset; Topspins, Ann Arbor, Mich) with one-way valves for automatic switching from gadopentetate dimeglumine injection to saline solution flush.
Image Analysis and Interpretation
To obtain reformation and maximum intensity projection images, the three-dimensional MR angiographic data were analyzed on a computer workstation. All MR angiograms were reviewed independently by three MR radiologists (R.C.C., Q.D., M.R.P.). The reviewers were blinded to gadopentetate dimeglumine concentration, but information was provided about which arm had been injected.
Signal intensity within the subclavian vein ipsilateral to the site of injection during the arterial phase was scored: score of 0, low; 1, intermediate; 2, high. The subclavian artery ipsilateral to the site of intravenous injection during the arterial phase was scored for the presence of the susceptibility artifact: score of 0, absent; 1.5, mild; 3, moderate; 4.5, severe. When MR angiograms are reviewed, subclavian arterial susceptibility artifact is more problematic than is high signal intensity in the subclavian vein. To weight the arterial artifact more heavily in the assessment of overall quality of an MR angiogram, the scoring system for subclavian arterial artifact had a wider range and interval size than that used for subclavian venous signal intensity. As a summary measure of the quality of a study, a combined score was obtained by adding the scores for subclavian arterial susceptibility artifact and subclavian venous signal intensity. The resultant score ranged between 0 and 6.5, with 12 unique possible scores, with a lower score corresponding to a higher quality study. The score corresponded to unique combinations of artifact and signal intensity levels to determine the relative quality of each study and was determined by consensus among the reviewers. For example, a study with high subclavian venous signal intensity and mild subclavian arterial artifact (combined score, 3.5) was considered better than a study with intermediate venous signal intensity and moderate arterial artifact (combined score, 4.0).
Statistical Analysis
For results with 33% versus 100% concentrations, the median combined scores for subclavian arterial artifact and subclavian venous signal intensity during the arterial phase were compared for each reviewer by using the Wilcoxon rank sum test. The 
statistic was used to analyze interobserver agreement for the combined scores with 33% and 100% concentrations but not with 25% concentration, since there was total agreement between all reviewers with this concentration.
For purposes of a statistical analysis and a meaningful comparison, a combined score of 3 or lower was considered to represent an acceptable MR angiographic study and a combined score greater than 3, unacceptable. A combined score of 3 or lower ensured that the subclavian arterial artifact was not more than moderate in severity. Studies with a score greater than 3 were considered unacceptable because an erroneous diagnosis could result from either the subclavian artery being obscured by overlapping bright vein or a subclavian artery being falsely described as moderately stenotic.
|
Results
|
|---|
Phantom
With the gradient-echo sequence, the artifact area adjacent to the test tubes increased with increasing concentration and echo time (Figs 2, 3a). Low signal intensity was seen in and around each test tube with high concentration of gadopentetate dimeglumine. High signal intensity was seen within only those tubes with low concentration. The concentration at which the tube turned from low to high signal intensity depended on the echo time. At an echo time of 10 msec, this transition occurred at a concentration of 5%; at an echo time of 1 msec, it occurred at a concentration of 30% (Figs 2, 3b).
View larger version (34K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3a. (a) Graph depicts artifact area for echo time (TE) plotted as a function of percentage concentration. Artifact increases with increasing echo time and concentration. (b) Graph depicts signal intensity plotted as a function of percentage concentration. Notice increased signal intensity with low concentration, especially at echo time of 1 msec (TE=1).
|
|
View larger version (30K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3b. (a) Graph depicts artifact area for echo time (TE) plotted as a function of percentage concentration. Artifact increases with increasing echo time and concentration. (b) Graph depicts signal intensity plotted as a function of percentage concentration. Notice increased signal intensity with low concentration, especially at echo time of 1 msec (TE=1).
|
|
To completely eliminate the artifact for the full range of echo times studied, the required concentration of gadopentetate dimeglumine was 1% or less. This concentration was not practical for use at MR angiography, however, because a large volume of dilute gadopentetate dimeglumine would have to be injected at an excessively fast rate. The injection rate for 5% concentration would have to be 20 times faster than that for nondilute (100%) gadopentetate dimeglumine. Since we currently use an injection rate of at least 1 mL/sec for nondilute gadopentetate dimeglumine, a 20:1 dilution would require a 20 mL/sec injection rate, which is not feasible. Furthermore, the subclavian vein would be bright during the arterial phase with extremely low concentration and might obscure the adjacent artery on maximum intensity projection images. With intermediate concentrations of 25% or 33%, the artifact was noted to decrease substantially with very short echo times (approximately 1 msec), but the contrast material could still be injected at a reasonable rate (3–4 mL/sec) while the subclavian vein remained dark from r2 effects. Accordingly, 25% and 33% concentrations were selected for further evaluation in patients.
Patients
Table 1 summarizes data for subclavian arterial susceptibility artifact and subclavian venous brightness with 25%, 33%, and 100% concentrations.
View this table:
[in this window]
[in a new window]
|
TABLE 1. Subclavian Arterial Artifact and Subclavian Venous Brightness on the Basis of Reviewer and Concentration of Gadopentetate Dimeglumine
|
|
With 25% concentration, subclavian arterial susceptibility artifact was completely eliminated in all patients, but high signal intensity was seen in the subclavian vein during the arterial phase (Fig 4a). Because of this problem, only the first three patients were examined with this dilution. The combined score was 2 for each reviewer, which indicates acceptable studies (Tables 1, 2).
View larger version (150K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4a. Maximum intensity projection images from arterial phase, coronal, three-dimensional, gadolinium-enhanced MR angiograms of the subclavian artery (SA) were obtained with (a) 25% concentration with right arm injection (5/1, 45° flip angle, 62.5-kHz bandwidth), (b) 33% concentration with right arm injection (5.69/1.1, 45° flip angle, 62.5-kHz bandwidth), and (c) 100% concentration with left arm injection (6.7/1.4, 45° flip angle, 62.5-kHz bandwidth). In a-c, SV = subclavian vein, SVC = superior vena cava. In a, susceptibility artifact at the right subclavian artery is completely eliminated, although there is high signal intensity in the superior vena cava and right subclavian vein ipsilateral to the site of injection. In b, the right subclavian vein demonstrates low signal intensity with only minimal susceptibility artifact (open arrow) in the subclavian artery. Also notice the signal void (solid arrow) at the region where the right subclavian vein drains into the superior vena cava. The signal dropout is related to increasing dilution of the concentrated gadopentetate dimeglumine as it mixes with blood from the jugular vein. In c, the left subclavian artery (open arrow) ipsilateral to the injection site demonstrates signal dropout from susceptibility artifact. In addition, high signal intensity (solid arrow) in the contralateral subclavian vein is due to reflux of contrast material.
|
|
View larger version (144K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4b. Maximum intensity projection images from arterial phase, coronal, three-dimensional, gadolinium-enhanced MR angiograms of the subclavian artery (SA) were obtained with (a) 25% concentration with right arm injection (5/1, 45° flip angle, 62.5-kHz bandwidth), (b) 33% concentration with right arm injection (5.69/1.1, 45° flip angle, 62.5-kHz bandwidth), and (c) 100% concentration with left arm injection (6.7/1.4, 45° flip angle, 62.5-kHz bandwidth). In a-c, SV = subclavian vein, SVC = superior vena cava. In a, susceptibility artifact at the right subclavian artery is completely eliminated, although there is high signal intensity in the superior vena cava and right subclavian vein ipsilateral to the site of injection. In b, the right subclavian vein demonstrates low signal intensity with only minimal susceptibility artifact (open arrow) in the subclavian artery. Also notice the signal void (solid arrow) at the region where the right subclavian vein drains into the superior vena cava. The signal dropout is related to increasing dilution of the concentrated gadopentetate dimeglumine as it mixes with blood from the jugular vein. In c, the left subclavian artery (open arrow) ipsilateral to the injection site demonstrates signal dropout from susceptibility artifact. In addition, high signal intensity (solid arrow) in the contralateral subclavian vein is due to reflux of contrast material.
|
|
View larger version (145K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4c. Maximum intensity projection images from arterial phase, coronal, three-dimensional, gadolinium-enhanced MR angiograms of the subclavian artery (SA) were obtained with (a) 25% concentration with right arm injection (5/1, 45° flip angle, 62.5-kHz bandwidth), (b) 33% concentration with right arm injection (5.69/1.1, 45° flip angle, 62.5-kHz bandwidth), and (c) 100% concentration with left arm injection (6.7/1.4, 45° flip angle, 62.5-kHz bandwidth). In a-c, SV = subclavian vein, SVC = superior vena cava. In a, susceptibility artifact at the right subclavian artery is completely eliminated, although there is high signal intensity in the superior vena cava and right subclavian vein ipsilateral to the site of injection. In b, the right subclavian vein demonstrates low signal intensity with only minimal susceptibility artifact (open arrow) in the subclavian artery. Also notice the signal void (solid arrow) at the region where the right subclavian vein drains into the superior vena cava. The signal dropout is related to increasing dilution of the concentrated gadopentetate dimeglumine as it mixes with blood from the jugular vein. In c, the left subclavian artery (open arrow) ipsilateral to the injection site demonstrates signal dropout from susceptibility artifact. In addition, high signal intensity (solid arrow) in the contralateral subclavian vein is due to reflux of contrast material.
|
|
With 33% concentration, the combined score was 3 or less for each reviewer, which indicates acceptable studies (Table 2). Mild subclavian arterial artifact was seen in some patients, but minimal or no high signal intensity was seen in the subclavian vein in the majority. Analysis of interobserver agreement for the eight subjects with 33% concentration yielded a value of 0.32 (P < .01), which indicates a fair level of agreement. With 33% concentration, each of the reviewers rated eight of eight studies as having an acceptable level of artifact, with a combined score of 3 or less (mean combined score from all reviewers, 1.54 on a scale of 0.0–6.5). These scores were statistically significantly superior to those with nondilute gadopentetate dimeglumine (mean combined score from all reviewers, 3.3).
With nondilute gadopentetate dimeglumine (100% concentration), the three reviewers scored acceptable studies in not more than half the patients (Table 2). Two reviewers scored acceptable studies in only three of the eight patients. Analysis of interobserver agreement for the nondilute gadopentetate dimeglumine studies yielded a value of 0.40 (P < .01). For all three reviewers, median scores were lower with 33% concentration than those with 100% concentration; hence, quality was higher with 33% concentration (reviewer 1, P = .05; reviewer 2, P = .1; reviewer 3, P = .1).
|
Discussion
|
|---|
Three-dimensional spoiled gradient-echo imaging during the arterial phase of gadopentetate dimeglumine injection has revolutionized MR angiography. However, gradient-echo imaging is highly sensitive to T2* effects. The paramagnetic effect of gadopentetate dimeglumine can produce local field inhomogeneities that cause spin dephasing on gradient-echo images. This spin dephasing may extend over several voxels that reach some distance from the concentrated gadopentetate dimeglumine. The effect is sometimes referred to as "susceptibility artifact." Normally, gadopentetate dimeglumine is sufficiently dilute by the time it reaches the aorta that there is no concentration-related artifact. However, artifact that obscures the subclavian artery has been observed when the concentration of gadopentetate dimeglumine in the subclavian vein ipsilateral to the site of injection is high. This artifact is transient and disappears as the concentration in the subclavian vein becomes more dilute in the equilibrium phase. The artifact may not be present with sequential mapping of k space if the gadopentetate dimeglumine bolus is administered in a short time and is followed by a large saline solution flush to clear the gadopentetate dimeglumine from the subclavian vein before central k-space data are acquired. The artifact is particularly problematic, however, when data are acquired with centric ordering of k space. Collection of the central k-space data at the beginning of imaging (centric ordering of k space) increases the likelihood that the center of k space is collected before the highly concentrated gadopentetate dimeglumine has cleared from the veins.
In this study, the data from an agar phantom and 19 patients demonstrated that this subclavian arterial artifact behaves like a susceptibility artifact. The subclavian artifact was reduced by diluting the gadopentetate dimeglumine and shortening the echo time. The pure T2 shortening effect of the gadopentetate dimeglumine also contributes to signal intensity reduction. However, this T2 effect should be confined to the subclavian vein and should not contribute to artifact that extends beyond the subclavian lumen. A benefit of this T2 effect is a reduction of subclavian venous signal intensity ipsilateral to the intravenous injection site. This reduction is particularly desirable on maximum intensity projection images to avoid the artery becoming obscured by overlapping vein. However, the associated susceptibility effect extending beyond the vein becomes undesirable when it blooms to encompass the adjacent subclavian artery. Thus, the gadopentetate dimeglumine dilution must be titrated so that the subclavian venous signal intensity is eliminated but the adjacent subclavian artery is unaffected. On the basis of our findings, we recommend use of a 33% concentration of gadopentetate dimeglumine with an echo time of 1 msec, the shortest echo time with our magnet.
The optimal goal is to dilute the gadopentetate dimeglumine just enough to prevent subclavian venous susceptibility artifact from affecting the adjacent subclavian artery while preserving sufficient T2 shortening of subclavian venous signal so the vein remains dark during arterial phase MR angiography. But even with 33% dilution, low subclavian venous signal intensity was not obtained in all patients. This finding presumably reflects varying degrees of further dilution of the injected contrast material by blood within the subclavian vein.
Because the degree of dilution varies from patient to patient, consistent suppression of venous signal intensity with complete elimination of subclavian arterial artifact is not possible at an echo time of 1 msec. As MR imagers become faster and achieve even shorter echo times, however, use of higher concentrations may attain more consistent suppression of subclavian venous signal intensity while avoiding the arterial artifact. Future availability of gadopentetate dimeglumine-based contrast agents with higher r1 and r2 relaxivity will also help mitigate this problem (17).
With use of slower magnets with longer minimum echo times, even greater dilution of gadopentetate dimeglumine will be required. This may not always be practical, because a larger fluid volume would have to be injected within the same period of time. For example, to dilute 20 mL of gadopentetate dimeglumine to a 10% concentration, 180 mL of 0.9% normal saline solution is needed for a total volume of 200 mL. This 200-mL volume would have to be injected at a rate of 10 mL/sec to complete administration of the bolus in 20 seconds. This rate is not acceptable in most patients. We found that the 33% (threefold) dilution of gadopentetate dimeglumine produces an acceptable level of artifact reduction at an echo time of 1 msec. Consequently, the injection rate must be increased threefold to deliver the same total dose of gadopentetate dimeglumine and necessitates the placement of a large caliber angiocatheter (
18 gauge). In addition, substitution of a larger volume of dilute gadopentetate dimeglumine might degrade MR angiographic studies overall compared with those obtained with a smaller volume of concentrated agent. We did not evaluate this substitution in our study.
To obtain the shortest possible echo time with our magnet, the bandwidth was maximized and thick partitions, a large field of view, and fractional echoes were used. Even shorter effective echo time may be possible with spiral k-space trajectories. A spiral sequence with a substantially shorter effective echo time may improve the signal-to-noise ratio and allow use of higher concentrations of contrast agent. The faster injection rates required for the dilute gadopentetate dimeglumine can affect the contrast material travel time (ie, the time for the contrast material to travel from the site of intravenous injection to the aorta). In this study, the problem of variation in contrast material travel time was avoided by using an automatic contrast material detection scheme to ensure synchronization of central k-space data acquisition with the arterial phase of the bolus.
Another way to eliminate this susceptibility artifact is to inject a compact bolus immediately followed by a large volume of saline solution flush timed so that the vein is filled with saline solution when the central k-space data are acquired. This can be done by using a gadolinium-enhanced three-dimensional fast MR angiographic sequence that allows shorter acquisition time and multiphase imaging (18) as well as sequential mapping of k space so that the central, low spatial frequency lines of k space are not acquired at the beginning of the study. Elevation of the patient’s arm may facilitate keeping the bolus compact, although this may lead to positional stenosis, another artifact that causes signal dropout over the subclavian artery. Monitoring of the bolus with MR fluoroscopy (19,20) may facilitate identification of the optimal moment to acquire central k-space data when the contrast material is in the subclavian artery and just clearing from the subclavian vein ipsilateral to the injection site. Use of body or phased-array coils with higher signal-to-noise ratio allows reduction of the total dose of contrast material so that it can be injected more quickly. Another strategy might be to use a central venous catheter that allows direct injection into the superior vena cava. Alternatively, a vein in a lower extremity could be used. Use of smaller voxels would also help reduce the susceptibility artifact, because larger voxels increase the likelihood that signal intensity will be canceled by out-of-phase spins. Use of smaller voxels, however, would increase the imaging time, since thinner partitions and more phase-encoding steps are required.
|
FOOTNOTES
|
|---|
Author contributions: Guarantors of integrity of entire study, M.A.N., M.R.P.; study concepts, M.R.P.; study design, M.A.N.; definition of intellectual content, M.R.P.; literature research, M.A.N.; clinical studies, M.A.N.; experimental studies, M.A.N., F.J.L.; data acquisition, M.A.N., F.J.L., Q.D., M.R.P., R.C.C., T.L.C.; data analysis, M.A.N.; statistical analysis, M.A.N.; manuscript preparation, M.A.N., M.R.P.; manuscript editing, T.L.C., Q.D., M.R.P.; manuscript review, all authors.
|
REFERENCES
|
|---|
-
Prince MR. Gadolinium-enhanced MR aortography. Radiology 1994; 191:155-164.[Abstract/Free Full Text]
-
Prince MR, Grist TM, Debatin JF. 3D contrast MR angiography Heidelberg, Germany: Springer, 1999.
-
Maki JM, Chenevert TL, Prince MR. 3D contrast enhanced MR angiography. Top Magn Reson Imaging 1996; 8:322-344.[Medline]
-
Goldfarb JW, Edelman RR. Coronary arteries: breath-hold, gadolinium-enhanced, 3D MR angiography. Radiology 1999; 206:830-834.[Abstract/Free Full Text]
-
Wielopolski PA, van Geuns RJ, de Feyter PJ, Oudkerk M. Breath-hold coronary MR angiography with volume-targeted imaging. Radiology 1998; 209:209-219.[Abstract/Free Full Text]
-
Meaney JF, Weg JG, Chenevert TL, Stafford-Johnson D, Hamilton BH, Prince MR. Diagnosis of pulmonary embolism with magnetic resonance angiography. N Engl J Med 1997; 336:1422-1427.[Abstract/Free Full Text]
-
Ho KY, Leiner T, de Hann MW, Kesels AG, Kitslaar PH, van Engelshoven JM. Peripheral vascular tree stenoses: evaluation with moving-bed infusion-tracking MR angiography. Radiology 1998; 206:683-692.[Abstract/Free Full Text]
-
Krinsky GA, Reuss PM, Lee VS, Carbognin G, Rofsky NM. Thoracic aorta: comparison of single-dose breath-hold and double-dose non–breath-hold gadolinium-enhanced three-dimensional MR angiography. AJR Am J Roentgenol 1999; 173:145-150.[Abstract/Free Full Text]
-
Huston J, III, Fain SB, Riederer SJ, Wilman AH, Bernstein MA, Busse RF. Carotid arteries: maximizing arterial to venous contrast in fluoroscopically triggered contrast-enhanced MR angiography with elliptic centric view ordering. Radiology 1999; 211:265-273.[Abstract/Free Full Text]
-
Korosec FR, Frayne R, Grist TM, Mistretta CA. Time-resolved contrast-enhanced 3D MR angiography. Magn Reson Med 1996; 36:345-351.[Medline]
-
Steiner P, McKinnon GC, Romanowski B, Geohde SC, Hany T, Debatin J. Contrast enhanced, ultrafast 3D pulmonary MR angiography in a single breathhold: initial assessment of imaging performance. J Magn Reson Imaging 1997; 7:177-182.[Medline]
-
Alley MT, Shifrin RY, Pelc NJ, Herfkens RJ. Ultrafast contrast-enhanced three-dimensional MR angiography: state of the art. RadioGraphics 1998; 18:273-285.[Abstract]
-
Prince MR, Chenevert TL, Foo TKF, Londy FJ, Ward JS, Maki JH. Contrast enhanced abdominal MR angiography: optimization of imaging delay time by automating the detection of contrast material arrival in the aorta. Radiology 1997; 203:109-114.[Abstract/Free Full Text]
-
Mezrich R. Prospective on k space. Radiology 1995; 195:297-315.[Free Full Text]
-
Maki JH, Prince MR, Londy FJ, Chenevert TL. The effects of time varying intravascular signal intensity signal and k-space acquisition order on three-dimensional MR angiography image quality. J Magn Reson Imaging 1996; 6:642-651.[Medline]
-
Lee YJ, Chung TS, Joo JY, Chien D, Laub G. Suboptimal contrast-enhanced carotid MR angiography from the left brachiocephalic venous stasis. J Magn Reson Imaging 1999; 10:503-509.[Medline]
-
Grist TM, Korosec FR, Peters DC, et al. Steady-state and dynamic MR angiography with MS-325: initial experience in humans. Radiology 1998; 207:539-544.[Abstract/Free Full Text]
-
Schoenberg SO, Bock M, Knopp MV, et al. Renal arteries: optimization of three-dimensional gadolinium-enhanced MR angiography with bolus-timing independent fast multiphase acquisition in a single breath hold. Radiology 1999; 201:667-679.
-
Riederer SJ, Tasciyan T, Farzaneh F, Lee JN, Wright RC, Herfkens RJ. MR fluoroscopy: technical feasibility. Magn Reson Med 1988; 8:1-15.[Medline]
-
Wilman AH, Riederer SJ, King BF, Debbins JP, Rossman JP, Ehman RL. Fluroscopically triggered contrast-enhanced three-dimensional MR angiography with elliptical centric view order: application to the renal arteries. Radiology 1997; 205:137-146.[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
F. Saremi, J. D. Grizzard, and R. J. Kim
Optimizing Cardiac MR Imaging: Practical Remedies for Artifacts
RadioGraphics,
July 1, 2008;
28(4):
1161 - 1187.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
K. Nael, M. Fenchel, M. Krishnam, G. Laub, J. P. Finn, and S. G. Ruehm
High-Spatial-Resolution Whole-Body MR Angiography with High-Acceleration Parallel Acquisition and 32-Channel 3.0-T Unit: Initial Experience
Radiology,
March 1, 2007;
242(3):
865 - 872.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
K. Nael, J. P. Villablanca, W. B. Pope, T. O. McNamara, G. Laub, and J. P. Finn
Supraaortic Arteries: Contrast-enhanced MR Angiography at 3.0 T--Highly Accelerated Parallel Acquisition for Improved Spatial Resolution over an Extended Field of View
Radiology,
February 1, 2007;
242(2):
600 - 609.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
X.-M. Zhang, D. G. Mitchell, M. Dohke, G. A. Holland, and L. Parker
Pancreatic Cysts: Depiction on Single-Shot Fast Spin-Echo MR Images
Radiology,
May 1, 2002;
(2002)
2232010815.
[Abstract]
[Full Text]
|
|
|
|
|
|
|
B. Tombach, C. Bremer, P. Reimer, F. Matzkies, R. M. Schaefer, W. Ebert, V. Geens, J. Eisele, and W. Heindel
Using Highly Concentrated Gadobutrol as an MR Contrast Agent in Patients Also Requiring Hemodialysis: Safety and Dialysability
Am. J. Roentgenol.,
January 1, 2002;
178(1):
105 - 109.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
X.-M. Zhang, D. G. Mitchell, M. Dohke, G. A. Holland, and L. Parker
Pancreatic Cysts: Depiction on Single-Shot Fast Spin-Echo MR Images
Radiology,
May 1, 2002;
223(2):
547 - 553.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
