AJR 2005; 185:750-755
© American Roentgen Ray Society
Contrast-Enhanced Bolus-Chased Whole-Body MR Angiography Using a Moving Tabletop and Quadrature Body Coil Acquisition
Darren D. Brennan,
Ciaran Johnston,
Julie O'Brien,
David H. Taylor,
Carmel Cronin and
Stephen J. Eustace
Department of Radiology, Cappagh National Orthopedic Hospital, Cappagh
Hospital, Finglas, Dublin, Ireland 11.
Received December 23, 2003;
accepted after revision November 19, 2004.
Address correspondence to S. J. Eustace
(Seustace{at}iol.ie).
Abstract
OBJECTIVE. MR angiography is now an established technique for the
evaluation of various arterial structures. With recent developments in
hardware, whole-body imaging has become technically feasible. The aim of this
study is to describe a technique for whole-body MR angiography using a
quadrature body coil and a moving tabletop.
CONCLUSION. Whole-body MR angiography is technically feasible and
may offer a method of screening for unsuspected disease.
Introduction
Whole-body MRI has undergone continuous refinement since its
earliest description and is now an accepted diagnostic tool for the assessment
of skeletal metastases [1].
Whole-body MR angiography has recently been introduced using prototypic
surface-coil acquisition [2].
We describe whole-body MR angiography that combines the extended field of view
afforded by a moving tabletop and the ease of use of quadrature body coil
detection. It is a technique that requires minimal patient preparation and
reduced room time and promises to offer an effective method of screening for
total-body atherosclerotic burden.
Subjects and Methods
Subjects
We recruited six volunteers and four symptomatic patients for our study.
Images were acquired on a 1.5-T system (Gyroscan Intera release 7, Philips
Medical Systems) with high-performance gradients (maximum amplitude, 30 mT/m;
slew rate, 75 mT/m/msec). The table was also equipped with a moving tabletop
extender (Mobi-Trak, Philips Medical Systems). No specific patient preparation
was required. The patients had an 18G cannula inserted into the antecubital
fossa. Patients then placed their hands above their heads and were advanced
into the imaging unit feet first and supine.
MRI
Whole-body MR angiography as we practice it relies on acquiring four
consecutive imaging stations or stacks that are subsequently melded together
at a prototypic workstation (ViewForm, Philips Medical Systems). Images are
acquired in multiple coronal stations after the acquisition of a test bolus of
contrast material. Each slab is prescribed from a whole-body axial localizing
time-of-flight (TOF) scan that is acquired in 3 min. The first station covers
the thoracic aorta and great vessels. The second station extends to the common
iliac vessels and the third and fourth stacks cover the pelvic vasculature and
proximal femoral vessels and run-off, respectively. A precontrast mask is
acquired before gadolinium is administered. Mask parameters are identical to
our image acquisition parameters (TR/TE, 4.1/1.36; flip angle, 30°;
rectangular field of view, 450 mm; matrix size, 512 x 272). Sixty
contiguous coronal slices are acquired per station with a 2-mm gap.
In all cases, we used gadobutrol (Gadovist, Schering), a gadolinium chelate
with a strength of 1.0 mol/L. Each patient received 0.15-0.2 mmol/kg body
weight delivered in a controlled fashion over approximately 50 sec by manual
injection, followed by a 40 mL saline flush. The technique was timed using a
test-bolus method. For the test bolus, a 2D gradient-echo technique was used
(7.0/1.09; flip angle, 40°; reconstructed matrix size, 2.0 x 2.0
x 8.0). Sequential sagittal oblique imaging of the thoracic aorta was
performed every second after the administration of a 2 mL of gadobutrol test
dose. The time from the commencement of the test bolus injection to its
arrival at the top of the aortic arch was used to calculate the time delay
from the injection of the contrast material to the start of the scan. After
the injection of the contrast bolus, images were acquired using a modified 3D
TOF sequence. This is an out-of-phase gradient-echo scan in which intervoxel
phase cancellation in background stationary tissues and ultrashort TRs produce
effective background signal suppression. Imaging parameters were 4.1/1.36;
flip angle, 30°; rectangular field of view, 450 mm; and matrix size, 512
x 272. Sixty contiguous coronal slices were acquired per station with a
2-mm gap. Automated voxel interpolation was used to yield reconstructed voxel
size of 0.88 x 0.88 x 2 mm. During image acquisition, four
integrated and automated table movements occur, allowing coverage of the whole
extracranial, extracardiac arterial tree. The initial two stations were
acquired using a 25-sec breath-hold for each station to avoid motion artifact.
A total scanning time of 87 sec can be achieved. The raw data were then
exported to an offline workstation for mask subtraction.
View larger version (36K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1 —Schematic of k-space filling using contrast-enhanced
timing-robust angiography technique. Random filling of predefined central
circle occurs that allows entire arterial bolus to be mapped to center of
k-space, maximizing contrast resolution. Remainder of k-space is filled during
arteriovenous window in concentric profile order, maximizing venous
suppression. (Fig. 1 courtesy of Koert Bloemers, Philips Medical Systems.)
|
|
View larger version (36K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2 —Mean quantitative and qualitative assessment of individual
arterial segments. SNR = signal-to-noise ratio (blue bars), R =
right, ICA = internal carotid artery, VA = vertebral artery, L = left, CCA =
common carotid artery, BCT = brachiocephalic trunk, LSCA = left subclavian
artery, DAA = (descending) abdominal aorta, DTA = descending thoracic artery,
AA = abdominal aorta, RA = renal artery, CIA = common iliac artery, CFA =
common femoral artery, SFA(P) = proximal superficial femoral artery, SFA(D) =
distal superficial artery/popliteal artery, PTA = posterior tibial artery, ATA
= ascending thoracic aorta, PA = peroneal artery. For ease of presentation,
mean qualitative score (red bars) recorded has been multiplied by
100.
|
|
View larger version (37K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3A —28-year-old female volunteer. Whole-body MR angiogram shows
good delineation of extracranial and extracardiac arterial system. There is
obvious jugular and portal venous overlay and filling of pulmonary vessels.
Right subclavian artery has been excluded from field of view due to technical
error.
|
|
In MR angiography, the method of k-space mapping is crucial. Most elliptic
techniques dictate that the center of k-space be filled at the exact same time
as the maximum contrast bolus, but the contrast-enhanced timing-robust
angiography k-space technique that we used during this technique is more
robust, especially in the third and fourth stages of the examination, where
inevitable dilution of contrast occurs. In this technique of randomly
segmented k-space ordering, a circle of variable diameter is defined, which
has its center in the very center of k-space (k 0, 0). During peak arterial
flow, random filling of this predefined central area occurs that allows
maximum usage of the upslope and peak areas of the arterial bolus, thus
optimizing contrast resolution. Subsequent definition is supplied by late
mapping of the edges of k-space by a more conventional concentric technique.
The diameter of the predefined circle influences the proportion of the
arterial bolus that is mapped to the center of k-space.
(Fig. 1). This method contrasts
with the more conventional linear, elliptical, or centric techniques used in
MR angiography that require the center of k-space to coincide exactly with the
peak arterial flow.
Image Interpretation
After transfer to the ViewForm workstation, the images were subsequently
melded together by a semiautomated process that was achieved in seconds.
Maximum intensity projection (MIP) was used to render projectional images
similar to conventional angiography. In addition, both source and rendered
images can be rotated, magnified, and interrogated in a number of planes on
the workstation.
View larger version (74K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3C —28-year-old female volunteer. Zoomed
maximum-intensity-projection images of great vessels allow easy depiction of
anatomy. These images can be interactively interrogated at workstation.
|
|
View larger version (44K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4 —64-year-old woman with history of endometrial carcinoma.
Whole-body MR angiography confirms vascular nature of two metastatic deposits
to lateral thigh compartment. No other vascular metastases are seen. Entire
extracranial and extracardiac vascular tree is well visualized. Left
subclavian is mostly excluded due to positioning.
|
|
We divided the arterial tree into 26 different segments in each patient and
performed quantitative and qualitative analysis on each segment. Segments 1
and 2 denoted the two internal carotid arteries; 3 and 4, the two vertebral
arteries; 5 and 6, the two common carotid arteries; 7, the brachiocephalic
trunk; 8, the left subclavian artery; 9, the ascending aorta; 10, the
descending thoracic aorta; 11, the abdominal aorta; 12 and 13, the renal
arteries; 14 and 15, the common iliac arteries; 16 and 17, the common femoral
arteries; 18 and 19, the proximal superficial femoral arteries; 20 and 21, the
distal superficial femoral/popliteal arteries; 22 and 23, the anterior tibial
arteries; 24 and 25, the posterior tibial arteries; and 26 and 27, the
peroneal arteries. For each arterial segment, we performed quantitative
analysis by calculating signal-to-noise ratio (SNR). This was done by manually
drawing a line (range, 2.3-40 mm; mean, 7 mm) on a segment of the artery of
interest and deriving mean signal with SD, from which the SNR was calculated.
All segmentation was done by one author. In addition, all arterial segments
were individually assessed by two authors experienced in MR angiography, in
consensus. A qualitative analysis of each arterial segment was performed
according to a four-point scale. In this scale, a score of 3 represented ideal
visualization of a vessel throughout its length. A score of 2 represented a
vessel that was visualized throughout its length but in which visualization
was sub-optimal. A score of 1 represented a vessel that was only partially
visualized throughout its length and deemed unsuitable for reliable
interpretation, and a score of 0 represented nonvisualization of a vessel. A
combination of source images and MIP images were used for examination of the
vessels.
View larger version (38K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5A —56-year-old man with symptomatic right-sided pain and history
of cemented hip arthroplasty 3 years previously. Whole-body MR angiogram
depicts long segment of stenosis in right common femoral artery, with
reconstitution in proximal right superficial artery. Venous contamination of
the run-off on left side is present, but reference to source images (not
shown) allows easy examination of these veins, except for posterior tibial
vein, which is partially obscured. Again, part of subclavian vein is excluded
due to positioning or possibly T2* effect from adjacent subclavian
vein.
|
|
View larger version (78K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5B —56-year-old man with symptomatic right-sided pain and history
of cemented hip arthroplasty 3 years previously. Composite images from
selective right-sided digital subtraction angiography confirm occlusion.
Patient's prosthesis does not interfere with MR images.
|
|
Results
A total of 260 segments were available for interrogation.
Figure 2 shows the mean SNR
calculated per segment. The highest SNRs were calculated, predictably, in the
aorta and central great vessels, with a gradual deterioration in the smaller
and more peripheral vessels, which could be predicted on the basis of contrast
dilution. Because the SNR is calculated by dividing the square root of the SD
into the mean calculated signal, a predictable drop could also be expected in
smaller peripheral vessels. Because image analysis depends on defining a
linear area of signal and because these defined areas are small in peripheral
vessels, they are more vulnerable to statistical fluctuations. Despite the
recorded drop in mean SNR, there was good delineation of the individual
arterial segments (Figs. 3A,
3B,
3C and
4). The mean calculated
qualitative scores are also shown in Figure
2. Of the 260 segments available, only three segments scored a
value of 0, two due to atherosclerotic occlusion (Figs.
5A,
5B and
6) and the other due to
nonvisualization of the right renal artery in a healthy volunteer. Doppler
interrogation of that renal artery was normal, and the MRI findings were
interpreted as a false positive.
View larger version (47K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 6 —48-year-old man with left-sided claudication. Whole-body MR
angiography shows marked irregularity of origin of left common iliac artery
(arrowhead). Long segment occlusion of mid superficial femoral artery
is also seen (arrow).
|
|
With the timing protocol outlined here, there were constant sources of
venous overlap. In all cases, early filling of the internal jugular veins
occurred, but reference to the source images or oblique MIP images allowed
easy visualization of the carotid and vertebral vessels. Similarly, in the
anteroposterior plane, the pulmonary arteries obscured much of the thoracic
aorta, but oblique images and source images allowed easy visualization of the
central arteries. The renal veins, portal vein, and proximal inferior vena
cava filled on most occasions, but in no case did this prevent visualization
of the adjacent renal arteries. Similar to conventional cases of peripheral
angiography, early venous return occasionally occurred, but reference to
source images allowed adequate delineation of the run-off, except in one case
where an otherwise well-enhanced posterior tibial artery was partially
obscured by early filling of its companion vein. Overall, visualization of the
vessels was high. In our symptomatic cohort, we were unable to detect other
significant foci of atherosclerotic disease, but conventional angiography,
when obtained in the symptomatic cases, was concordant with the whole-body
angiograms. Reference standard digital subtraction angiography was not
available in any other patient, as these were asymptomatic volunteers assumed
to have normal vasculature. The proximal subclavian artery on the right side
(three cases) and left side (one case) was obscured from vision in four cases
due to positioning, which was initially performed with the patients in
abduction and external rotation (hand with cannula above head), but
repositioning with their hands by their sides corrected this problem.
Discussion
Imaging has represented the gold standard for diagnosis of atherosclerosis
since the widespread introduction of percutaneous angiography. Initial
enthusiasm for peripheral angiography has been tempered over time by the
physical complications of intervention in tandem with toxicity imposed by the
contrast materials used. Noninvasive attempts at assessment of the vascular
tree led to the development of Doppler sonography. However, because this
system is operator dependent and limited by patient habitus, alternatives have
been sought. MR angiography appears to represent the solution. Because it is
noninvasive and nonionizing, objective MR angiography has been widely embraced
and adopted. It is largely independent of body habitus and only the largest of
patients cannot be examined. Using gadolinium chelates, which are inert
compounds, avoids many of the nephrotoxic and allergic reactions encountered
with iodinated media. Although it is technically involved, consistently
good-quality angiograms can be obtained with appropriate technologist
training. Similarly, reviewer training is necessary so that the images can be
consistently and correctly interpreted. Since its introduction by Prince et
al. [3] in 1993, technical
refinements such as bolus-chase techniques, faster gradients, and improvements
in coil detection have placed MR angiography as the test of choice for
suspected vascular disease in most institutions
[4]. The refinement of 3D MR
angiography has made near isotropic voxel acquisition possible.
Although MR angiography has been well established in the targeted
assessment of the various manifestations of atherosclerosis including
renovascular disease, carotid artery disease, and peripheral vascular disease,
technical limitations in hardware and contrast agents have precluded the
development of a whole-body MR assessment. The development of such a strategy
is logical, as a wealth of epidemiological evidence points to atherosclerosis
being a systemic disease. As such, a systemic approach seems logical, as
pointed out by Goyen et al.
[5]. Although their strategy
differs somewhat from ours, many similarities exist. There is little doubt
that the use of surface coils improves resolution, but we have offset some of
this expected signal loss by a combination of high-strength gadolinium, which
has already been proven to be effective, and judicious use of k-space mapping,
where the entire arterial bolus is mapped to the center of k-space.
Paramagnetic contrast agents are pivotal to contrast-enhanced MR angiography
[6], as the ultrashort
acquisition times preclude signal return from nonenhanced tissue. Gadobutrol
1.0 is a macrocyclic neutral gadolinium compound in which the physical
properties allow delivery of twice the dose of gadolinium without adversely
increasing osmolality or the excellent safety profile of lower strength
gadolinium [7]. It has already
shown promise in perfusion imaging in the brain; peripheral MR angiography;
and, more recently, whole-body MR angiography
[5]. The increased SNR
[8] offsets the expected loss
in signal from the absence of a surface coil.
Some limitations are associated with this technique. First, the
intracranial vessels are rarely seen, although these are rarely the source of
atherosclerotic pathology. However, the coronary arteries are not seen, and
epidemiological evidence suggests a high association between peripheral
vascular disease and coronary artery disease. Second, the technique requires
two 25-sec breath-hold examinations, which can be difficult in some
populations. Third, our patient population was small and did not include
patients with critical ischemia, which traditionally represent the more
difficult patient population to image. Although manual injection was used in
our study (we do not have a pump), we believe that the consistently good
results obtained with such a technique reflect its versatility rather than act
as detraction. It is logical to assume controlled contrast bolus delivery with
a pump would provide equivalent if not better results. Finally, these studies
were done with hardware (a moving tabletop) not available to everyone, and the
images were reconstructed on a prototypic software platform not available to
the general public. However, the source images could be analyzed on any
available platform or workstation and the whole-body MIP images, while useful
to display overall disease importance, are probably of secondary importance
because constant sources of venous overlap mandate recourse to the source
images frequently. This inevitably increases the time necessary for image
interpretation.
In conclusion, we have shown the feasibility of performing whole-body MR
angiography using a quadrature body coil, high-strength gadolinium, and a
moving tabletop. This allows accurate visualization of the entire
extracranial, extracardiac arterial system without patient preparation and
minimum additional room time (mean increase in room time over peripheral MR
angiography, 5.5 min). The technique has the potential to provide noninvasive
assessment of the entire arterial system, thus allowing early detection of
asymptomatic foci of atherosclerotic vascular disease.
References
- Eustace S, Tello R, DeCarvalho V, et al. A comparison of whole-body
turboSTIR MR imaging and planar 99mTc-methylene diphosphonate scintigraphy in
the examination of patients with suspected skeletal metastases.
AJR 1997; 169:1655
-1661[Abstract/Free Full Text]
- Goyen M, Herborn CU, Kroger K, Lauenstein TC, Debatin JF, Ruehm SG.
Detection of atherosclerosis: systemic imaging for systemic disease with
whole-body three-dimensional MR angiography—initial experience.
Radiology 2003;227
: 277-282[Abstract/Free Full Text]
- Prince MR, Yucel EK, Kaufman JA, Harrison DC, Geller SC. Dynamic
gadolinium-enhanced three-dimensional abdominal MR arteriography. J
Magn Reson Imaging 1993; 3:877
-881[Medline]
- Goyen M, Ruehm SG, Debatin JF. MR angiography for assessment of
peripheral vascular disease. Radiol Clin North Am2002; 40:835
-846[CrossRef][Medline]
- Goyen M, Herborn CU, Vogt FM, et al. Using a 1 M Gd-chelate
(gadobutrol) for total-body three-dimensional MR angiography: preliminary
experience. J Magn Reson Imaging 2003;17
: 565-571[CrossRef][Medline]
- Goyen M, Ruehm SG, Debatin JF. MR-angiography: the role of contrast
agents. Eur J Radiol 2000;34
: 247-256[CrossRef][Medline]
- Balzer JO, Loewe C, Davis K, et al. Safety of contrast-enhanced MR
angiography employing gadobutrol 1.0 M as contrast material. Eur
Radiol 2003; 13:2067
-2074[CrossRef][Medline]
- Goyen M, Lauenstein TC, Herborn CU, Debatin JF, Bosk S, Ruehm SG.
0.5 M Gd chelate (Magnevist) versus 1.0 M Gd chelate (Gadovist):
dose-independent effect on image quality of pelvic three-dimensional
MR-angiography. J Magn Reson Imaging2001; 14:602
-607[CrossRef][Medline]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
Abstract
Introduction
