DOI:10.2214/AJR.04.1130
AJR 2005; 185:1069-1073
© American Roentgen Ray Society
Contrast-Enhanced MR Angiography of Subclavian Steal Syndrome: Value of the 2D Time-of-Flight "Localizer" Sign
Niall Sheehy1,
Sorcha MacNally1,
Clare S. Smith1,
Gerard Boyle1,
Prakash Madhavan2 and
James F. M. Meaney1
1 Department of Diagnostic Imaging, St. James's Hospital, James's St., Dublin 8,
Ireland.
2 Department of Vascular Surgery, St. James's Hospital, Dublin 8, Ireland.
Received July 18, 2004;
accepted after revision October 15, 2004.
Address correspondence to N. Sheehy
(niallsheehy{at}iolfree.ie).
Abstract
OBJECTIVE. Our objective was to determine if direction of flow
within the vertebral artery could be reliably determined by evaluation of
flow-sensitive, low-resolution 2D time-of-flight (TOF) localizer images taken
before 3D contrast-enhanced MR angiography (3D CEMRA) sequences in patients
with unsuspected subclavian steal syndrome.
CONCLUSION. Vertebral artery patency on 3D CEMRA in cases in which
the vessel is absent on the TOF localizer in association with ipsilateral
subclavian artery stenosis indicates reversal of flow in the vertebral artery
and confirms the subclavian steal phenomenon. The combination of anatomic
imaging with 3D CEMRA with functional information provided by the
low-resolution TOF localizer confirms the diagnosis of subclavian steal
without additional imaging.
Introduction
Three-dimensional contrast-enhanced MR angiography (3D CEMRA) has virtually
eliminated catheter arteriography for a variety of indications, including
evaluation of extracranial vasculature
[1] and great vessels
[2]. High-resolution arterial
phase images are generated by synchronizing central k-space data with the
arterial peak of a bolus of contrast agent delivered IV. Despite impressive
accuracy for detection and grading of carotid and vertebral artery stenosis,
this technique is limited in the diagnosis of subclavian steal syndrome, as it
does not encode direction of flow. The purpose of this study was to determine
if direction of flow within the vertebral artery could be determined from a
low-resolution time-of-flight (TOF) localizer routinely used for localizing
purposes in our department.
Subjects and Methods
Two groups of patients were assessed: In the first group, seven patients
(four men and three women; mean patient age, 62 years; age range, 49-84 years)
with subclavian steal and flow reversal within the vertebral artery confirmed
on Doppler sonography underwent 3D CEMRA.
The second group comprised 50 consecutive patients (32 men and 18 women;
mean age, 59 years; age range, 15-80 years) who had MRA performed for
assessment of carotid arteries over a 6-month period. Indications for MRA were
as follows: possible dissection (n = 4), carotid stenosis (n
= 36), confirmation of internal carotid occlusion (n = 6), subclavian
stenosis (n = 2), and postoperative assessment (n = 2).
MRI
All images were acquired on a 1.5-T MRI scanner equipped with fast
gradients (23 mT/m, rise time 200 µsec).
2D TOF Localizer
Scan parameters for the 2D TOF were TR/TE, 30/7; field of view, 260 x
220 mm; matrix, 256 x 128; number of signal averages (NSA), 1; 5-mm
slices with 5-mm gap (images were interpolated for
maximum-intensity-projection [MIP] images). Forty slices were acquired in the
axial scan plane. Acquisition time was 96 sec. Anteroposterior, lateral, and
craniocaudal MIPs were automatically generated to facilitate placement of the
subsequent bolus detection and 3D imaging volume. A traveling saturation pulse
was used to suppress venous flow.
Bolus Detection
A single thick 2D slice (70 mm) oriented in the coronal plane was placed
over the region of interest (TR/TE, 5/1.4; 256 x 192; NSA 1). The
initial image served as a mask for subtraction purposes, and subtracted images
were reconstructed and displayed in real time on the display monitor. Images
were acquired with a temporal resolution of one image/sec.
3D CEMRA
A fast-spoiled gradient-echo acquisition was prescribed from the localizing
sequence. The coronally oriented imaging volume was centered on the midline
and carefully tailored to give adequate coverage of the carotid and vertebral
arteries. Scan parameters were TR/TE, 4.7/1.7; flip angle, 25°; field of
view, 260 x 200; and matrix, 256 x 192. True voxel size was
approximately 1 mm3 and acquisition time was approximately 40
sec.
View larger version (61K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1A —63-year-old woman. Reformatted maximum-intensity-projection
(A) and sample axial image (B) from 2D time-of-flight localizer
in healthy patient undergoing 3D contrast-enhanced MR angiography at our
institution. Vertebral arteries are visible, indicating normal flow direction
and absence of severe vertebral disease.
|
|
View larger version (73K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1B —63-year-old woman. Reformatted maximum-intensity-projection
(A) and sample axial image (B) from 2D time-of-flight localizer
in healthy patient undergoing 3D contrast-enhanced MR angiography at our
institution. Vertebral arteries are visible, indicating normal flow direction
and absence of severe vertebral disease.
|
|
Twenty milliliters of gadolinium contrast agent (gadobenate dimeglumine 0.5
mol/L, MultiHance; Bracco) was administered using a biphasic injection rate
(1.0 mL/sec x 10 sec followed by 0.6 mL/sec for 17 sec for a total
injection duration of 27 sec).
The 3D acquisition was initiated by the operator once the contrast material
was visualized within the extracranial arteries on the 2D bolus-detection
sequence. Because of the relatively long scanning time, breath-holding was not
used, and the delay between aborting the 2D scan and initiating the 3D scan
was approx 1 sec. A recessed (1 sec) elliptic-centric k-space filling order
was used.
Image Evaluation
All images were interpreted by two independent radiologists blinded to the
clinical details. TOF images and 3D CEMRA were reviewed separately. TOF images
were reviewed for presence or absence of each vertebral artery on both axial
and reformatted images (Figs.
1A and
1B). The 3D CEMRA sequence was
reviewed for presence or absence of the vertebral artery on each side, and for
significant stenosis (> 50%) or occlusion of the subclavian artery proximal
to the take-off of the vertebral artery on both sides. In addition,
atherosclerotic disease of the brachiocephalic artery was assessed. Presence
or absence of vertebral artery disease was also evaluated on a four-point
scale. Differences in interpretation were resolved by consensus.
Results
In the first group, all patients with sonography visualization of flow
reversal in the vertebral artery appeared to have an occluded vertebral
artery, as evidenced by a flow void in the vertebral artery on the 2D TOF
localizer despite a normal, patent vertebral artery on 3D CEMRA. All patients
had ipsilateral subclavian/brachiocephalic stenosis or occlusion on 3D CEMRA
(Figs. 2A,
2B,
2C,
2D,
3A, and
3B).
View larger version (112K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2A —70-year-old woman with suspected subclavian steal syndrome
and flow reversal on Doppler sonography. Occlusion is seen within left
subclavian artery proximal to veterbral artery origin, shown on
high-resolution 3D contrast-enhanced MR angiography (CEMRA)
maximum-intensity-projection (MIP) image. The left vertebral artery appears
normal (A). The normal left vertebral artery can also be seen on axial
reformat of data (B). Corresponding reformatted MIP (C) and
axial (D) images from 2D time-of-flight localizer show flow void in
left vertebral artery, despite visualization of normal right vertebral artery
on 3D CEMRA. This indicates flow reversal within vertebral artery
("localizer" sign).
|
|
View larger version (37K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2B —70-year-old woman with suspected subclavian steal syndrome
and flow reversal on Doppler sonography. Occlusion is seen within left
subclavian artery proximal to veterbral artery origin, shown on
high-resolution 3D contrast-enhanced MR angiography (CEMRA)
maximum-intensity-projection (MIP) image. The left vertebral artery appears
normal (A). The normal left vertebral artery can also be seen on axial
reformat of data (B). Corresponding reformatted MIP (C) and
axial (D) images from 2D time-of-flight localizer show flow void in
left vertebral artery, despite visualization of normal right vertebral artery
on 3D CEMRA. This indicates flow reversal within vertebral artery
("localizer" sign).
|
|
View larger version (90K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2C —70-year-old woman with suspected subclavian steal syndrome
and flow reversal on Doppler sonography. Occlusion is seen within left
subclavian artery proximal to veterbral artery origin, shown on
high-resolution 3D contrast-enhanced MR angiography (CEMRA)
maximum-intensity-projection (MIP) image. The left vertebral artery appears
normal (A). The normal left vertebral artery can also be seen on axial
reformat of data (B). Corresponding reformatted MIP (C) and
axial (D) images from 2D time-of-flight localizer show flow void in
left vertebral artery, despite visualization of normal right vertebral artery
on 3D CEMRA. This indicates flow reversal within vertebral artery
("localizer" sign).
|
|
View larger version (58K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2D —70-year-old woman with suspected subclavian steal syndrome
and flow reversal on Doppler sonography. Occlusion is seen within left
subclavian artery proximal to veterbral artery origin, shown on
high-resolution 3D contrast-enhanced MR angiography (CEMRA)
maximum-intensity-projection (MIP) image. The left vertebral artery appears
normal (A). The normal left vertebral artery can also be seen on axial
reformat of data (B). Corresponding reformatted MIP (C) and
axial (D) images from 2D time-of-flight localizer show flow void in
left vertebral artery, despite visualization of normal right vertebral artery
on 3D CEMRA. This indicates flow reversal within vertebral artery
("localizer" sign).
|
|
View larger version (76K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3A —Another patient with proximal right subclavian artery
stenosis and normal right vertebral artery on 3D contrast-enhanced MR
angiography maximum intensity projection (A). Axial image from 2D
time-of-flight localizer again shows flow void in right vertebral artery,
indicating flow reversal (B).
|
|
View larger version (63K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3B —Another patient with proximal right subclavian artery
stenosis and normal right vertebral artery on 3D contrast-enhanced MR
angiography maximum intensity projection (A). Axial image from 2D
time-of-flight localizer again shows flow void in right vertebral artery,
indicating flow reversal (B).
|
|
In the second group, a flow void was seen in the vertebral artery in nine
patients. Two of these patients had subclavian stenosis and a patent vertebral
artery on CEMRA, indicating vertebral artery flow reversal that was
subsequently confirmed by Doppler sonography.
Of the remaining seven patients in whom a flow void was seen on the 2D TOF
localizer, the 3D CEMRA sequence revealed three patients with vertebral artery
occlusion, three with severe vertebral stenosis, and one with dissection
(Figs. 4A and
4B). Thus, all patients with a
flow void in the vertebral artery on the 2D TOF localizer when that vertebral
artery was not severely diseased on 3D CEMRA had reversal of flow on Doppler
sonography.
View larger version (105K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4A —Patient shows severe atherosclerotic stenosis of right
vertebral artery (A). 2D time-of-flight localizer shows flow void
(B), this time due to severe vertebral disease shown on 3D
contrast-enhanced MR angiography sequences.
|
|
View larger version (106K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4B —Patient shows severe atherosclerotic stenosis of right
vertebral artery (A). 2D time-of-flight localizer shows flow void
(B), this time due to severe vertebral disease shown on 3D
contrast-enhanced MR angiography sequences.
|
|
Discussion
In 1960, Contorni [3] coined
the term "subclavian steal" to describe the finding of reversal of
normal direction of flow in the vertebral artery ipsilateral to a severe
stenosis or occlusion occurring between the aortic arch and vertebral artery
origin. In 1961, Fisher [4]
noted the association of subclavian steal with neurologic symptoms, which he
named "subclavian steal syndrome," and noted that this phenomenon
may lead to brainstem ischemia and stroke, either at rest or after arm
exercise.
Although subclavian or innominate artery stenosis is not rare (occurring in
17% of 6,534 cases in the joint study of extracranial arterial occlusion
[5]) flow reversal within the
vertebral artery is present in a minority of these cases (2.5% in the same
study), and of those with angiographic steal, only 5.3% (9/168) had neurologic
symptoms. This emphasizes the point that neurologic symptoms do not always
accompany flow reversal within the vertebral artery, and for a neurologic
deficit to occur, other factors (e.g., impaired collateral blood supply to the
posterior circulation) must coexist. Therefore, differentiation between simple
flow reversal within the vertebral artery without associated neurologic
symptoms (subclavian steal phenomenon) and the syndrome of transient
neurologic symptoms related to cerebral ischemia (subclavian steal syndrome)
in patients with retrograde vertebral artery flow should be made.
In patients with subclavian steal, the mechanism by which exercise-induced
increased upper extremity blood-flow demand can be met is by diversion
("stealing") of blood into the subclavian artery distal to a
severe stenosis or occlusion by reversal of flow within the ipsilateral
vertebral artery. This phenomenon is possible because of the unique anatomic
disposition of the vertebral and subclavian arteries, as anastomosis of the
vertebral arteries to form the basilar artery provides a potential natural
conduit for bypass of an occluded subclavian artery (on the left) or
brachiocephalic artery (on the right). Therefore, in the presence of a
stenosis or occlusion proximal to the vertebral artery origin, increased upper
extremity blood flow demand is met by stealing of blood from the cerebral
circulation via retrograde vertebral artery flow. However, because of anatomic
variability, a different situation may exist on each side. On the left side,
for subclavian steal to occur, the vertebral artery must take its origin from
the subclavian artery; this occurs in 94% of subjects. In the remaining 6% of
subjects in whom the left vertebral artery arises directly from the aortic
arch, the patient is deprived of the collateral route that compensates for
reduced blood flow to the left arm and subclavian steal cannot occur. There is
no corresponding anatomic variability on the right (the vertebral artery
always arises from the subclavian artery). Therefore, a natural conduit is
always present that allows compensatory flow into the right subclavian artery
via the right vertebral artery.
The standard noninvasive method for diagnosis of subclavian steal
phenomenon is Doppler sonography. Although Doppler sonography can reliably
show flow reversal in the vertebral artery, visualization of the relevant
pathology within the subclavian artery is difficult or impossible, and most
patients are referred for further imaging. Catheter angiography has the
advantages of high intrinsic spatial resolution and high temporal resolution
and remains the gold standard for evaluation of disease of the great vessels,
including subclavian steal. The high temporal resolution allows clear and
unambiguous visualization of both the relevant stenosis/occlusion in
association with antegrade vertebral artery flow on the "normal"
side, retrograde filling of the vertebral artery, and late filling of the
subclavian artery on the "abnormal" side. This finding is evident
on aortic flush injection and, when performed, selective injection of the
contralateral vertebral artery. Catheter angiography was not performed on any
patient in our study, representing a limitation of our work.
Many clinicians, however, are reluctant to refer patients for catheter
angiography because of its attendant morbidity. MRA methods, many of which are
sensitive to direction of flow, offer an attractive alternative. For example,
phase-contrast MRA methods inherently encode direction of flow
[6] and can show subclavian
stenosis and reversal of flow in the vertebral artery
[7]. TOF MRA does not possess
inherent flow-encoded information but can show flow direction, as flow from
one direction can be suppressed by a saturation pulse that facilitates
generation of selective arterial or venous MR images
[8]. Although these techniques
are suited to visualization of flow reversal in subclavian steal phenomenon
[9], contrast-enhanced
techniques have largely replaced noncontrast techniques for evaluation of the
extracranial vessels in clinical practice. This is due to clear advantages of
3D contrast-enhanced methods, including higher signal-to-noise ratio, higher
contrast, and higher spatial resolution and shorter scan times
[10-12].
On the other hand, without the additional directional information provided
by other techniques, 3D CEMRA has a potential disadvantage in the evaluation
of patients with suspected subclavian steal syndrome as a standalone
technique. Although we have abandoned TOF MRA for diagnostic MRA, we use a
low-resolution TOF acquisition for the localizer that is performed before the
contrast-enhanced sequence. This is a technique that has been adopted by many
centers to ensure accurate tailoring of the 3D contrast-enhanced imaging
volume to the region of interest (to maximize resolution while minimizing
scanning time). This localizer sequence, performed with a cephalad traveling
saturation pulse to suppress venous flow, generates an arteriographic image
that, although clearly inadequate for diagnostic purposes, nonetheless clearly
demarcates the limits of the relevant arteries and allows accurate placement
of the 3D imaging volume. This study differs from a diagnostic TOF MRA only in
spatial resolution (low resolution is used to ensure a short scanning time)
and gives identical directional information as a higher resolution diagnostic
TOF MRA. Because of the saturation pulse, reversal of flow within a vertebral
artery will present as a flow void, thus confirming that a patent vertebral
artery on 3D CEMRA contains reversed flow, giving the same information as a
formal phase-contrast or TOF MRA sequence without additional time penalty. As
we have shown, if the vertebral artery is occluded or has severe stenosis, the
TOF localizer will also fail to visualize the vertebral artery. This
distinction between a flow void due to flow reversal and one due to a diseased
vertebral artery is easily made on the high-resolution 3D CEMRA sequence.
If the information on flow direction present on the TOF localizer sequence
is not assessed, then in cases in which subclavian steal is suspected before
the MR examination, a formal high-resolution TOF study or a phase-contrast
study must also be performed, at the cost of additional scanning time. In
cases where a question of possible subclavian steal syndrome is only raised in
the light of the 3D CEMRA sequence diagnosing unsuspected subclavian or
brachiocephalic stenosis, the patient must be recalled for an additional
procedure, either MR angiography or Doppler sonography.
In conclusion, the low-resolution TOF localizer, which is normally
discarded from the diagnostic evaluation, may indicate direction of vertebral
artery flow. We therefore recommend routine inspection of the 2D TOF localizer
sequence in all 3D CEMRA studies for discrepancy between visualization of the
vertebral artery on the TOF localizer and the 3D CEMRA, particularly in cases
in which patients have posterior circulation symptoms and may have clinically
unsuspected subclavian steal. We refer to this finding as the localizer sign
and have shown that it allows subclavian steal phenomenon to be diagnosed in
patients undergoing 3D CEMRA without requiring further investigation.
References
- Huston J III, Fain SB, Wald JT, et al. Carotid artery: elliptic
centric contrast-enhanced MR angiography compared with conventional
angiography. Radiology 2001;218
: 138-143[Abstract/Free Full Text]
- Randoux B, Marro B, Koskas F, et al. Proximal great vessels of
aortic arch: comparison of 3D gadolinium-enhanced MR angiography and digital
subtraction angiography. Radiology 2003;229
: 697-702[Abstract/Free Full Text]
- Contorni L. The vertebro-vertebral collateral circulation in
obliteration of the subclavian artery at its origin [in Italian].
Minerva Chir 1960;15
: 268-271[Medline]
- Fisher CM. A new vascular syndrome: "the subclavian
steal." New Engl J Med 1961;265
: 912-913
- Fields WS, Lemak NA. Joint study of extracranial arterial
occlusion. VII. Subclavian steal—a review of 168 cases.
JAMA 1972; 222:1139
-1143[Abstract/Free Full Text]
- Dumoulin CL. Phase contrast MR angiography techniques.
Magn Reson Imaging Clin N Am 1995;3
: 399-411[Medline]
- Turski PA, Levine R, Turnipseed W, Kennell T. MR angiography flow
analysis. Neurovascular applications. Magn Reson Imaging Clin N
Am 1995; 3:541
-555[Medline]
- Felmlee JP, Ehman RL. Spatial presaturation: a method for
suppressing flow artifacts and improving depiction of vascular anatomy in MRI.
Radiology 1987;164
: 559-564[Abstract/Free Full Text]
- Drutman J, Gyorke A, Davis WL, Turski PA. Evaluation of subclavian
steal with 2D phase-contrast and 2D time-of-flight MR angiography.
Am J Neuroradiol 1994;15
: 1642-1645[Abstract]
- Prince MR, Narasimham DL, Stanley JC, et al. Breath-hold
gadolinium-enhanced MR angiography of the abdominal aorta and its major
branches. Radiology 1995;197
: 785-792[Abstract/Free Full Text]
- Cloft HJ, Murphy KJ, Prince MR, Brunberg JA. 3D gadolinium-enhanced
MR angiography of the carotid arteries. Magn Reson
Imaging 1996; 14:593
-600[CrossRef][Medline]
- 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]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
Abstract
Introduction
