DOI:10.2214/AJR.07.3665
AJR 2008; 191:1734-1739
© American Roentgen Ray Society
Low Injection Rate for 3D Moving-Table Bolus-Chase MR Angiography: Initial Experience with 3-T Imaging to Allay Venous Contamination in the Calf
Kai Lin1,2,
Zhao-Qi Zhang1,
Jun-Yan Sun1,
Zhan-Ming Fan1 and
Biao Lu1
1 Department of Radiology, Beijing Institute of Heart, Lung, and Blood Vessel
Diseases and Beijing Anzhen Hospital, Capital Medical University, Beijing,
China.
2 The Russell H. Morgan Department of Radiology and Radiological Science,
Division of Neuroradiology, The Johns Hopkins University School of Medicine,
600 N Wolfe St., Phipps B-110, Baltimore, MD 21287-2182.
Received January 11, 2008;
accepted after revision July 8, 2008.
Abstract presented at the 2007 European Congress of Radiology, Vienna,
Austria (tracking no. 07-P-515-ECR).
Address correspondence to K. Lin
(kai{at}jhmi.edu).
Abstract
OBJECTIVE. The purpose of this study was to evaluate the potential
for improving the image quality of 3D bolus-chase peripheral MR angiography by
injecting contrast medium at a slow rate.
SUBJECTS AND METHODS. Using similar imaging parameters in all cases,
we performed bolus-chase MR angiography of the abdominal and lower limb
arteries of 80 patients. The injection protocol for 40 patients had three
parts: 20 mL of gadopentetate dimeglumine at 2 mL/s, 8 mL of gadopentetate
dimeglumine at 1 mL/s, and 20 mL of saline solution at 1 mL/s. For the other
40 patients, the injection protocol was 20 mL of gadopentetate dimeglumine at
1.2 mL/s, 8 mL of gadopentetate dimeglumine at 0.7 mL/s, and 20 mL of saline
solution at 0.7 mL/s. Using independent Student's t tests between
groups, we compared signal-to-noise and contrast-to-noise ratios in the
abdomen and pelvis, the thigh, and the calf. Arterial visibility and venous
contamination on 3D images of the calf were graded and compared.
RESULTS. The lower injection rate increased arterial visibility
(p < 0.001), reduced venous contamination in the calf (p
< 0.001), and increased the contrast-to-noise ratio in the calf (p
< 0.001). At the upper levels, signal-to-noise and contrast-to-noise ratios
did not differ between the two groups.
CONCLUSION. At 3-T MRI, a lower injection rate may alleviate venous
contamination and increase arterial visibility in the calf while
signal-to-noise and contrast-to-noise ratios at higher levels are
maintained.
Keywords: 3 T • contrast medium • injection protocol • MR angiography • peripheral arterial occlusive disease
Introduction
As a first-pass MR angiographic technique, multilevel 3D boluschase
peripheral MR angiography has facilitated the evaluation of peripheral
vascular disease because its accuracy is comparable with that of traditional
catheter angiography [1]. In
general, with moving-table bolus-chase procedures, images of the abdominal
aorta, pelvis, and lower extremities are obtained consecutively. This method
allows rapid imaging on a large bodily scale with a single injection of
gadolinium contrast agent. The patient is advanced through the MRI unit while
the contrast agent flows down the peripheral arteries. Thus, the same contrast
bolus can be imaged several times in the abdomen and pelvis, thigh, and
calf.
An optimal vascular MRI examination should be performed after the contrast
medium arrives in the target artery but before it dissipates into the veins.
In many patients, however, optimal imaging is not feasible because the MRI
unit cannot match the rate of flow of the gadolinium bolus through the calves
[2]. In these patients,
contrast medium is not distributed appropriately, and venous enhancement
occurs in the distal aspect, that is, the calves.
A slower rate of injection of contrast medium is expected to alleviate
venous contamination because the delay of peak enhancement in the arteries
results in a longer arteriovenous transit time and a delay of peak enhancement
in the veins [3,
4]. However, a slow injection
rate can adversely affect the signal intensity of enhancement owing to low
concentrations of contrast medium in the vessel
[3]. The aim of this study was
to balance venous contamination and arterial signal intensity. We tested the
hypothesis that with a high-field-strength (3 T) system, a slow injection rate
can alleviate contamination of veins in the calves while maintaining image
quality at the two upper levels.
Subjects and Methods
Patients
As approved by the institutional review board, 80 consecutively registered
patients (57 men, 23 women; age range, 39–78 years; mean body weight,
66.9 kg; range, 50–78 kg) with symptomatic peripheral arterial occlusive
disease (PAOD) were recruited for bolus-chase three-level MR angiography of
the abdominal and lower-limb arteries. Symptomatic PAOD was diagnosed by a
surgeon (20 years of experience) using the combination of a less than 0.95
ratio between the systolic pressure at the ankle divided by the highest
systolic arm pressure and the presence of intermittent claudication in at
least one limb [5]. All
patients were randomly assigned to group A (40 patients) or group B (40
patients). Informed consent was obtained in all cases. Fontaine stages I, II,
III, and IV were identified in all 160 limbs. The condition of 57 limbs
(35.6%) was Fontaine stage I (asymptomatic); 67 limbs (41.9%), Fontaine stage
II (intermittent claudication); 35 limbs (21.9%), Fontaine stage III (rest
pain); and one limb (0.6%), Fontaine stage IV (trophic lesion). In group A,
there was one case of slight cellulitis of one thigh and one case of wet
gangrene. Two ulcers were found in group B. The patient characteristics are
shown in Table 1.
General MRI Parameters
All imaging was performed with a 3-T MRI system (Signa Excite, GE
Healthcare) with a gradient amplitude of 40 mT x m–1
and a maximum slew rate of 150 T x m–1 x
s–1. One operator performed all of the examinations using
similar imaging parameters (Table
2). The MR angiographic pulse sequence was triggered according to
the time of peak enhancement of the test bolus at the origin of the renal
artery, which was determined through analysis of the signal intensity curve of
the test bolus determined with the software of the MRI system. A real-time
test bolus was administered with MR fluoroscopy (matrix size, 128 x 96).
In group A, the test bolus was 2 mL of contrast agent administered at 2 mL/s
followed by 20 mL of saline solution administered at 1 mL/s. In group B, the
test bolus was 2 mL of contrast agent administered at 1.2 mL/s followed by 20
mL of saline solution administered at 0.7 mL/s.
Injection Protocol
The contrast medium was gadopentetate dimeglumine (0.5 mmol/mL, Magnevist,
Bayer Schering Pharma). The injection rate and dose were set on a power
injector. All contrast media was injected through a 22-gauge integrated
catheter placed into the dorsal vein of the left hand. The injection protocol
for group A (40 patients) had three parts: 20 mL of gadopentetate dimeglumine
at 2 mL/s, 8 mL of gadopentetate dimeglumine at 1 mL/s, and 20 mL of saline
solution at 1 mL/s. Group B (40 patients) received an updated injection
protocol of 20 mL of gadopentetate dimeglumine at 1.2 mL/s, 8 mL of
gadopentetate dimeglumine at 0.7 mL/s, and 20 mL of saline solution at 0.7
mL/s.
Image Evaluation
All images were processed on a dedicated workstation (Advantage
Workstation, software version 4.2, GE Healthcare). Signal-to-noise ratio
(SNR), defined as mean signal intensity in the imaging field divided by the SD
of signal intensity outside the imaging field, was measured at the origins of
the renal artery, superficial femoral artery, and posterior tibial artery. If
the posterior tibial artery was occluded, we chose the anterior tibial artery
(three limbs) and the peroneal artery (one limb) for analysis.
Contrast-to-noise ratio (CNR) was defined as the difference in SNRs of the
abdominal aorta and liver, femoral artery and quadriceps femoris muscle, and
posterior tibial artery and gastrocnemius muscle.
Subjective evaluation and measurements were done by two radiologists (6 and
8 years of experience) who worked independently and were blinded to the
protocol used. Arterial visibility and venous contamination were assessed
separately. Venous contamination on 3D images of the calf was ranked 0, none;
1, trace; 2, mild; 3, moderate; 4, severe. General arterial visibility in the
calf was evaluated for diagnostic quality on a modified 5-point scale
[6]: 0, not seen; 1, barely
seen, cannot exclude many pathologic conditions; 2, seen but with compromise
of diagnosis; 3, seen with almost no compromise of diagnosis; 4, well seen,
definite diagnosis without interference.
Statistical Analysis
The average SNR and CNR from different imaging levels were compared between
groups with the independent Student's t test. Arterial visibility and
venous contamination in the calf were compared for the two injection protocols
(independent Student's t test) and the Fontaine stages with one-way
analysis of variance. SPSS software (version 10.0, SPSS) was used to analyze
the data. A value of p < 0.05 was considered to indicate a
statistically significant difference.
View larger version (60K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2 —46-year-old man with Fontaine stage III peripheral arterial
occlusive disease on right side and Fontaine stage II disease on left.
Injection protocol was 2 mL of contrast agent and 1 mL of saline solution at 1
mL/s. MR angiogram shows veins enhanced in calves. Arterial visibility score
is 2 on right side and 3 on left. Venous contamination score is 3 on both
sides. Calf contrast-to-noise ratio is 26.6 on right side and 27.8 on left.
Decrease in signal intensity in femoral artery may be due to dielectric
effect. 
|
|
Results
From the beginning of injection, complete arterial phase image acquisition
lasted approximately 100 seconds. One patient (a man in group B, two limbs,
Fontaine stage III) was excluded because the protocol failed owing to mistimed
triggering of image acquisition. In group A, a limb with Fontaine stage IV
disease was excluded from analysis because of poor visibility of the main
vessels in the calf. Therefore, a total of 79 cases (157 limbs, 79 in group A
and 78 in group B) were available for analysis. The entire workflow and
timetable of crucial control points are shown in
Figure 1. Typical images
obtained with the two injection protocols are shown in Figures
2 and
3.
View larger version (40K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3 —54-year-old man with Fontaine stage II peripheral arterial
occlusive disease on both sides. Injection protocol was 1.2 mL of contrast
agent and 0.7 mL of saline solution at 0.7 mL/s. Contrast-enhanced MR
angiogram shows right external iliac artery and left femoral artery are
occluded. Arterial visibility score is 4 on both sides. Venous contamination
score is 0 on both sides. Calf contrast-to-noise ratio is 38.2 on right side
and 36.9 on left. Decrease in signal intensity in femoral artery may be due to
dielectric effect. 
|
|
Compared with group A, group B had a longer injection duration (57 vs 38
seconds) and a higher CNR in the calf (31.7 vs 22.4, p < 0.001).
There was no statistical difference between the two groups in SNR at any level
and no statistical significance in CNR in the abdominal aorta and thigh. The
quantitative evaluation for two upper levels is shown in
Table 3.
There was a high level of agreement between the two readers on subjective
evaluation of arterial visibility (
= 0.758) and venous contamination
( = 0.812). The venous contamination score of group A was higher than
that of group B (2.46 vs 1.05, p < 0.001, for reader 1; 2.29 vs
0.95, p < 0.001, for reader 2). Conversely, the arterial
visibility score was higher in group B than in group A (3.24 vs 2.56,
p < 0.001, for reader 1; 3.12 vs 2.51, p < 0.001, for
reader 2). There was no statistical difference in either arterial visibility
or venous contamination score among the Fontaine stages within the same groups
(Tables 4 and
5).
Discussion
The moving-table bolus-chase 3D contrast-enhanced MR angiographic technique
has been accepted as effective in the evaluation of peripheral runoff vessels
and is being widely used [1].
According to the principles of MR angiography, imaging relies on the passage
of contrast material through the target vessel during image acquisition rather
than being redistributed in the body. If acquisition begins well before bolus
arrival, the signal intensity is weak because of the small amount of contrast
material in the target vessel. Conversely, if acquisition begins too long
after bolus arrival, the accompanying veins or tissues become strongly
enhanced and interfere with the diagnosis. Therefore, it is crucial to
pinpoint a suitable time window for image acquisition. In bolus-chase
multilevel MR angiography the timing of acquisition can be optimized only for
the first level (abdomen and pelvis). Regardless of the time of bolus arrival,
image acquisition at the thigh and calf is scheduled with a prearranged
program and is unalterable after image acquisition is initiated.
Prince et al. [7] found that
the mean travel time of contrast material to the common femoral artery was 24
± 6 seconds with an additional 5 ± 2 seconds to reach the
popliteal artery and 7 ± 4 seconds to reach the tibial arteries. In our
study, the three consecutive image acquisitions at the three levels began 23
± 1 seconds, 47 ± 3 seconds, and 70 ± 4 seconds after
injection. Although there may be a difference between the times of bolus
arrival and peak enhancement, which depends on different methods of
measurement, it is clear that contrast medium travels faster than the MRI
table. In the thigh, image acquisition begins several seconds after the
arrival of contrast material. In the calf, the time gap is so much greater
that contrast medium appears in the veins once image acquisition begins. For
this reason, image quality is generally better for evaluation of the first two
levels (abdomen and pelvis, thigh), and quite often it is inadequate for
assessment of the last level (calves and feet)
[8–10].
Because there is an incongruity between table movement and bolus transit,
synchronizing these two entities becomes a great challenge to operators.
Numerous technologic advances have been made to address the time gap. Use of
an outer compress reduces venous return
[2,
11,
12], and parallel imaging
techniques, simultaneous acquisition of spatial harmonics
[13] and sensitivity encoding
[14] with a multichannel coil
[15], and the scoutless
stepping-table peripheral contrast-enhanced MR angiographic technique
[16] save imaging time at the
two upper levels. Another effective method, hybrid peripheral MR angiography
[6,
17,
18], also may produce images
free of venous contamination. Hybrid technique does not have to synchronize
the bolus and the moving bed; the calf and upper level are imaged separately,
and two independent MR angiographic processes are combined.
The key point of our technique is to prolong the arterial to venous
transition because short transit times result in venous signal intensity in
the calf [19]. We carefully
balanced conflicts in bolus transit time, bed movement, and appropriate signal
intensity in a solid bolus-chase procedure. With a low injection rate, which
has not been formally recommended in the manuals of MRI units, the 3-T MRI
system maintained signal strength with a relatively low concentration of
contrast material. At the same time, the T2* effect of the slow
injection rate [20] and lower
background signal intensity, including that of veins, may contribute to
increased CNR in the arteries of the calf.
A higher-strength magnetic field is considered an immediate source for
better image quality, especially for higher SNR and greater spatial
resolution, the traditional indexes. In our study, we indirectly traded high
magnetic field strength for low venous contamination by decreasing the
injection rate. To our knowledge, this attempt is the first to reduce venous
contamination in 3D bolus-chase peripheral MR angiography by taking advantage
of a high magnetic field strength.
There were limitations to this pilot study. First, most of the subjects had
slight or moderate PAOD (Fontaine stage I or II). A major limitation was lack
of a significant number of cases of severe disease (Fontaine stage III and
IV). Most of the patients with severe PAOD directly underwent imaging with the
reference standard, digital subtraction angiography, at our hospital because
they could not keep their limbs stable during MR angiography because of acute
pain. Because evaluation of the accuracy of different imaging techniques was
not the aim of this study, to avoid unnecessary examinations, we skipped this
traditional comparison and clinical evaluation of MR angiographic findings.
Further experiments are needed before this method can be successfully applied
to patients with severe PAOD.
A second limitation of the study was that vascular lesions such as PAOD
greatly affect bolus travel time
[21]. However, with injection
protocol in this study, we did not find that Fontaine stages are related to
the severity of venous contamination, most likely because the effects of
abnormal blood flow are overwhelmed by asynchronism between image acquisition
and bolus transit. Nevertheless, we need more cases and a well-designed
clinical trial to explore the relation between severity of PAOD and degree of
venous contamination. Third, we did not measure the exact bolus traveling
time, as other researchers have done
[7,
21]. We believe doing so would
have required extra injections in clinical patients. Nevertheless, the better
image quality we observed allowed us to infer that we might have achieved our
intended goal.
A fourth limitation was that in addition to bolus transit time,
abnormalities such as arteriovenous shunt, fistula, and collateral circulation
can affect venous contamination. We did not take these abnormalities into
account because there was no suitable quantitative model to cover the various
conditions comprehensively. Fifth, the experimental injection rate used in the
study was based on our daily experience but not on experimental data. Because
of the limitations of clinical examination, we did not determine an optimal
injection protocol for all patients. We thus used feasible rates, 1.2, 0.7,
and 0.7 mL/s, to test the existence of a compromise between injection rate and
signal intensity in a high-strength magnetic field (3 T).
With a 3-T system, use of a slow injection rate can alleviate venous
contamination and increase arterial CNR in the calf without sacrificing SNR
and CNR at the two upper levels. As such, high magnetic field strength
unconventionally contributes to the image quality of 3D bolus-chase peripheral
MR angiography in a simple but technically valid approach.
Acknowledgments
We thank the medical and technical staff members of the MRI department,
especially Hong Jiang, Miao Guo, and Haixia Yang, for their expert technical
assistance. We appreciate the technical support of Nan Sun, whose employment
status at GE Healthcare (China) did not influence the data in this study. We
thank Anthony Portanova for his assistance in editing this manuscript.
References
- Meaney JF, Ridgway JP, Chakraverty S, et al. Stepping-table
gadolinium-enhanced digital subtraction MR angiography of the aorta and lower
extremity arteries: preliminary experience. Radiology1999; 211:59
-67[Abstract/Free Full Text]
- Zhang H, Ho B, Mohajer K, Xu H, Peripheral magnetic resonance
angiography with a multicompartment curved leg wrap for thigh compression.
J Cardiovasc Magn Reson 2007;9
: 659-664[Medline]
- Bae KT. Peak contrast enhancement in CTA and MRA: when does it
occur and why? Radiology 2003;227
: 809-816[Abstract/Free Full Text]
- Herold T, Paetzel C, Völk M, et al. Contrast-en hanced
magnetic resonance angiography of the carotid arteries: influence of injection
rates and volumes on arterial-venous transit time. Invest
Radiol 2004; 39:65
-72[Medline]
- Stoffers HE, Rinkens PE, Kester AD, Kaiser V, Knottnerus JA. The
prevalence of asymptomatic and unrecognized peripheral arterial occlusive
disease. Int J Epidemiol 1996;25
: 282-290[Abstract/Free Full Text]
- Tongdee R, Narra VR, McNeal G, et al. Hybrid peripheral 3D
contrast-enhanced MR angiography of calf and foot vasculature.
AJR 2006; 186:1746
-1753[Abstract/Free Full Text]
- Prince MR, Chabra SG, Watts R, et al. Contrast material travel
times in patients undergoing peripheral MR angiography.
Radiology 2002;224
: 55-61[Abstract/Free Full Text]
- Foo TK, Ho VB, Hood MN, Marcos HB, Hess SL, Choyke PL.
High-spatial-resolution multistation MR imaging of lower extremity peripheral
vasculature with segmented volume acquisition: feasibility study.
Radiology 2001;219
: 835-841[Abstract/Free Full Text]
- Wang Y, Winchester P, Khilnani M, et al. Contrast-enhanced
peripheral MR angiography from the abdominal aorta to the pedal arteries:
combined dynamic two-dimensional and bolus-chase three-dimensional
acquisitions. Invest Radiol 2001;36
: 170-177[CrossRef][Medline]
- Hany TF, Carroll TJ, Omary RA, et al. Aorta and runoff vessels:
single injection MR angiography with automated table movement compared with
multiinjection time-resolved MR angiography— initial results.
Radiology 2001;221
: 266-272[Abstract/Free Full Text]
- Bilecen D, Schulte AC, Bongartz G, et al. Infragenual
cuff-compression reduces venous contamination in contrast-enhanced MR
angiography of the calf. J Magn Reson Imaging2004; 20:347
-351[CrossRef][Medline]
- Herborn CU, Ajaj W, Goyen M, Massing S, Ruehm SG, Debatin JF.
Peripheral vasculature: whole-body MR angiography with midfemoral venous
compression—initial experience. Radiology2004; 230:872
-878[Abstract/Free Full Text]
- Sodickson D, Manning W. Simultaneous acquisition of spatial
harmonics (SMASH): fast imaging with radiofrequency coil arrays.
Magn Reson Med 1997;38
: 591-603[Medline]
- Pruessmann K, Weiger M, Scheidegger M, Boesiger P. SENSE:
sensitivity encoding for fast MRI. Magn Reson Med1999; 42:952
-962[CrossRef][Medline]
- Kramer H, Michaely HJ, Matschl V, et al. High-resolution magnetic
resonance angiography of the lower extremities with a dedicated 36-element
matrix coil at 3 Tesla. Invest Radiol2007; 42:477
-483[CrossRef][Medline]
- Li W, Vu AT, Tutton S, et al. Scoutless steppingtable peripheral
contrast-enhanced MR angiography. J Magn Reson Imaging2006; 23:235
-241[CrossRef][Medline]
- Morasch MD, Collins J, Pereles FS, et al. Lower extremity
stepping-table magnetic resonance angiography with multilevel contrast timing
and segmented contrast infusion. J Vasc Surg2003; 37:62
-71[CrossRef][Medline]
- Pereles FS, Collins JD, Carr JC, et al. Accuracy of stepping-table
lower extremity MR angiography with dual-level bolus timing and separate calf
acquisition: hybrid peripheral MR angiography.
Radiology 2006;240
: 283-290[Abstract/Free Full Text]
- Wang Y, Chen CZ, Chabra SG, et al. Bolus arterial-venous transit in
the lower extremity and venous contamination in bolus chase three-dimensional
magnetic resonance angiography Invest Radiol2002; 37:458
-463[CrossRef][Medline]
- de Bazelaire C, Rofsky NM, Duhamel G, et al. Combined
T2* and T1 measurements for improved perfusion and permeability
studies in high field using dynamic contrast enhancement. Eur
Radiol 2006; 16:2083
-2091[CrossRef][Medline]
- Fleischmann D, Rubin GD. Quantification of intravenously
administered contrast medium transit through the peripheral arteries:
implications for CT angiography. Radiology2005; 236:1076
-1082[Abstract/Free Full Text]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
Abstract
Introduction
