| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Radiology, NYU Medical Center, 530 First Ave., New York, NY
10016.
2 Present address: Department of Radiology, Beth Israel Deaconess Medical
Center, 330 Brookline Ave., Boston, MA 02215.
Received January 8, 2002;
revised May 15, 2002;
N. M. Rofsky receives research support from Berlex Laboratories.
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Twenty patients successfully completed both lower extremity MR angiography and digital subtraction angiography within a 3-month period. For MR angiography, patients were placed on the scanner table on a standard plastic patient-transfer board. We performed unenhanced and contrast-enhanced imaging at the level of the pelvis using a three-dimensional gradient-echo sequence (TR range/TE range, 3.8-4.6/1.3-1.8; flip angle range, 25-40°). Then patients were quickly pulled 350-400 mm using the transfer-board handles, and two subsequent acquisitions were obtained at the level of the thighs. For each modality, two radiologists who were unaware of correlative imaging results retrospectively scored all vessel segments as either greater than or equal to 50% stenosis or less than 50% stenosis, and interobserver agreement was determined. Using digital subtraction angiography as the standard of reference, we used consensus data to compute MR angiography sensitivity and specificity.
RESULTS. In the 261 vessel segments considered, MR angiography had a
sensitivity of 75% (12/16) and a specificity of 98% (94/96) for the detection
of stenosis greater than or equal to 50% from the aorta through the common
femoral arteries. For the superficial and profunda femoral arteries through
the popliteal arteries, these values were 97% (31/32) and 94% (34/36),
respectively. MR angiography interobserver agreement for detection of stenosis
was good (
= 0.68) for the aorta through the common femoral arteries
and excellent ( = 0.88) for the superficial and profunda femoral
arteries through the popliteal arteries. These values were comparable to those
found for digital subtraction angiography ( = 0.67 and = 0.88,
respectively).
CONCLUSION. Stationary-table MR angiography is a useful, simple strategy for lower extremity angiography in centers without a moving table.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
One challenge of peripheral MR angiography is the need for broad anatomic coverage. For the evaluation of patients with claudication, inclusion of vascular anatomy from the infrarenal aorta through the popliteal arteries will often suffice, whereas for patients with clinical evidence of limb-threatening ischemia who are surgical candidates, additional distal coverage through the feet is necessary. Broad coverage has generally been accomplished by dividing the vascular territory of interest into multiple stations, each of which can be studied separately using contrast-enhanced MR angiography or flow-dependent techniques [3,4,5,6,7,8].
Recently, single-injection bolus-chase moving-table techniques have been reported for multistation lower extremity arterial imaging [9,10,11,12,13]. Such technology is new in its implementation and, as such, is not standardized. This technology requires MR imaging hardware and software features that many centers do not have, including MR unit tables that are freely movable between acquisitions [9,10,11,12] or a specialized moving platform that can be placed on a stationary MR unit table [13]. We have developed and implemented a technique that relies on manual patient translation using a standard plastic patient-transfer board and can thus be performed on any scanner system. The purpose of this study was to compare the utility of this stationary-table bolus-chase peripheral MR angiography strategy with conventional digital subtraction angiography (DSA).
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
MR Angiography
After we obtained written informed consent for the administration of the
contrast material, patients underwent imaging with a 1.5-T MR imaging system
(Vision [n = 15] or Symphony [n = 5]; Siemens, Erlangen,
Germany). MR imaging was performed at two contiguous anatomic stations in all
patients using a body coil. The extent of the first station was usually from
the infrarenal aorta through the common femoral arteries, and the second
station typically extended from the superficial and profunda femoral arteries
to the proximal calf vessels. At each station, coronal 3D contrast-enhanced MR
angiographic images were acquired using a T1-weighted, slice-interpolated 3D
gradient-echo sequence with linear sequential phase-encoding. The following
imaging parameters were used: TR range/TE range, 3.8-4.6/1.3-1.8; flip angle
range, 25-40°; field of view range, 425-475 mm; typical in-plane pixel
size of 2.1 mm (range, 1.6-2.6 mm) x 1.9 mm (range, 1.7-2.8 mm); typical
postinterpolation section thickness of 1.5 mm (range, 1.0-2.0 mm); average
acquisition time, 23 ± 4.6 sec. For each patient, all 3D acquisitions
used identical parameters. The field of view (from 425 to 475 mm) and the
distance each patient was moved (from 350 to 400 mm) were selected to achieve
approximately 75 mm of overlap; this ensured adequate coverage and allowed for
any lack of precision during the manual movement of the table.
A 22-gauge IV catheter was inserted in the antecubital vein and attached to an MR-compatible power injector (Spectris; MedRad, Pittsburgh, PA). The power injector was loaded with 40 mL of gadopentetate dimeglumine (Magnevist; Berlex Laboratories, Wayne, NJ) for each patient (average dose, 0.24 mmol/kg; range, 0.16-0.36 mmol/kg). Patients were placed on a plastic patient-transfer board feet-first in the supine position on the MR imaging table. They were initially positioned to enable imaging at the level of the pelvis. Adhesive distance markers were placed on the table and board to mark the desired interval of patient translation. Cushions were placed under the patients' knees and ankles to align the vessels of interest in the horizontal plane, and an elastic binder was used to secure the lower legs.
Unenhanced imaging was performed in the first station only (pelvis), to provide a template for subtraction from contrast-enhanced imaging in this station. For the second station, an unenhanced mask image was not obtained because adequate coregistration of the unenhanced and contrast-enhanced 3D acquisitions could not be ensured using the stationary-table method. After unenhanced images were obtained, a timing examination was performed at the level of the pelvis (iliac vessels) using a test dose of 1 mL of gadopentetate dimeglumine followed by 20 mL of saline solution, both injected at a rate of 2 mL/sec. Patient circulation time, as measured by the timing examination, was used to determine timing for contrast-enhanced MR angiography in the first station according to methods described in the literature [18, 19].
For MR angiography, after a bolus injection of the remaining 39 mL of gadopentetate dimeglumine at a rate of 2 mL/sec (bolus injection duration, 19.5 sec), three consecutive 3D data sets were acquired in quick succession, starting at the level of the pelvis with a 5-sec delay between each acquisition. During the first 5-sec delay, patients were quickly pulled the previously determined distance (350-400-mm) in the cephalad direction manually using the handles of the patient-transfer board, enabling the subsequent two acquisitions to be obtained at the level of the thighs. Two people were needed to perform the manual patient translation. The length of time each patient was on the table, including image reconstruction time, was approximately 30 min.
Digital Subtraction Angiography
All DSA examinations were performed by experienced angiographers using
commercially available equipment (Integris 4000; Philips Medical Systems,
Bothell, WA). Most patients (14/20) underwent DSA of the distal aorta; the
pelvis, including the proximal superficial femoral arteries; and a single
lower extremity distal in relation to the level of the foot. Three patients
underwent a single lower extremity run-off study with evaluation limited to
the common femoral artery through the foot on the side of interest. Three
patients underwent a complete bilateral run-off evaluation from the distal
aorta to the feet.
For imaging of the abdominal aorta through the common iliac bifurcation, a 5-French pigtail catheter (Angiodynamics, Queensbury, NY) was inserted into the suprarenal abdominal aorta under fluoroscopic guidance, and 20 mL of ioxaglate (19.6% sodium, 39.3% meglumine) (Hexabrix; Mallinckrodt, St. Louis, MO) was administered at 20 mL/sec. Subsequent evaluation of the pelvis (infrarenal aorta through proximal superficial femoral arteries) was performed after administration of an additional 16 mL of contrast material at 8 mL/sec. For single lower extremity run-off examinations, a 5-French endhole catheter (Angiodynamics) was advanced into the external iliac artery on the side of interest, and multistation imaging was performed from the common femoral artery to the level of the foot using 52-56 mL of contrast material administered in multiple increments at a rate of 4 mL/sec. For bilateral lower extremity examinations, all contrast material injections were administered through a 5-French pigtail catheter (Angiodynamics) inserted into the suprarenal abdominal aorta, and bilateral multistation imaging was performed distal in relation to the level of the feet. For all studies, multiple standard and oblique coronal views were obtained as necessary to adequately show all vessels of interest. Selected images from each patient were saved on optical disks with the intention of adequately showing all vessels of interest.
Image Analysis
Two radiologists with experience in MR angiography interpretation
independently evaluated each MR angiographic examination, and two vascular
radiologists independently evaluated each DSA examination. These reviewers
were unaware of the results of correlative studies. For the evaluation of MR
angiography, analysis of the first station (pelvis) used source images and
subtracted data in which unenhanced images were subtracted from
contrast-enhanced images, whereas analysis of the second-station data used
only source data. All cases were reviewed interactively on a commercially
available MR imaging workstation (Siemens) using both
maximum-intensity-projection and multiplanar reconstructions. DSA images were
reviewed using hard-copy radiographs created from images saved on optical
disks during initial DSA evaluations.
For analysis of vessel segments, the arterial vasculature was divided into 21 vessel segments, namely the infrarenal abdominal aorta and the right and left arteries of the common and external iliac arteries; the common, superficial, and profunda femoral arteries; the popliteal arteries; the tibioperoneal trunks; the proximal anterior and posterior tibial arteries; and the proximal peroneal arteries. Vessel segments were evaluated for adequacy of visualization and, if depiction was thought to be adequate, scored on the basis of whether the highest grade lesion produced greater than or equal to 50% stenosis or produced less than 50% stenosis. The radiologists who reviwed the MR angiograms and DSA images were made aware of the surgical history of the three patients in whom a peripheral arterial bypass graft (femoral—popliteal) was present and asked to evaluate whether the grafts were patent or occluded.
After independent review, same-modality discrepancies in vessel segments were resolved by consensus. If both radiologists agreed that visualization of a segment was inadequate for evaluation, the segment was excluded from both MR angiography and DSA statistical analysis.
Data Analysis
The data were analyzed both in terms of vessel segment type (e.g., common
femoral artery or superficial femoral artery) and in terms of vascular
territories corresponding to anatomy typically visualized in the first station
(infrarenal aorta through common femoral arteries), the proximal second
station (superficial and profunda femoral arteries through popliteal
arteries), and the distal second station (tibioperoneal trunks and proximal
tibial and peroneal arteries). To assess inter-observer agreement between
same-modality reviewers, we calculated the observed concordance (the number of
segments for a given vessel segment type that were assigned identical scores
by each reviewer divided by the total number of segments evaluated) and kappa
coefficients for each vessel segment type and for the specified vascular
territories. The interpretation of kappa coefficient values followed accepted
guidelines ( > 0.75 indicated excellent reproducibility, 0.4 <
0.75 indicated good reproducibility, and 0 < 0.4
indicated marginal reproducibility)
[20,
21]. We calculated the
sensitivity and specificity of MR angiography for each vessel segment and
vascular territory using DSA as the standard of reference. Ninety-five percent
confidence intervals (CI) were determined for all computed sensitivity and
specificity values on the basis of binomial distribution.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
For the remaining 261 vessels considered for data analysis, Table 1 presents the MR angiography sensitivity and specificity (with 95% CI) and the corresponding MR angiography and DSA kappa coefficients and concordance values for interobserver agreement for each vessel segment type.
|
We found that the stationary-table MR angiography sensitivity for detection of stenoses greater than or equal to 50% from the aorta through the common femoral arteries (usually corresponding to first-station imaging) was 75% (12/16; 95% CI, 0.48-0.93), and the specificity was 98% (94/96; 95% CI, 0.93-1.0) (Fig. 1A,1B). Corresponding MR angiography and DSA kappa coefficients were 0.68 and 0.67, respectively. For the superficial and profunda femoral arteries and popliteal arteries (usually corresponding to proximal second-station imaging), MR angiography sensitivity was 97% (31/32; 95% CI, 0.84-1.0), and specificity was 94% (34/36; 95% CI, 0.81-0.99), with MR angiography and DSA both yielding kappa coefficients of 0.88. For the tibioperoneal trunk and proximal anterior and posterior tibial and peroneal arteries (usually the distal second station), MR angiography sensitivity was 75% (18/24; 95% CI, 0.53-0.90), specificity was 89% (51/57; 95% CI, 0.78-0.96), and MR angiography and DSA kappa coefficients were 0.66 and 0.84, respectively (Fig. 2A,2B).
|
|
|
|
For the three patients with femoral—popliteal bypass grafts, all four radiologists independently found two grafts to be patent and one occluded.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Particularly encouraging is the correlation of MR angiography with DSA, despite the use of a body coil for all MR imaging and evaluation of second-station vessel segments without subtracted data. Ho et al. [9] reported sensitivity and specificity values ranging from 94% to 100% and 97% to 100%, respectively, for detection of stenotic disease greater than or equal to 50% in the superficial and profunda femoral arteries and in the popliteal artery when using a bolus-chase moving-table strategy with a body coil. Ruehm et al. [10] used a moving-table strategy with a dedicated peripheral vascular coil and obtained a sensitivity of 92% and specificity of 98% for detection of stenotic disease greater than or equal to 50% when all vessel segments from the aorta through the calf vessels were included. Meaney et al. [11] used a bolus-chase moving-table strategy with a body coil and a homemade clamp to assist in precise patient translation and reported a sensitivity and specificity of 81% and 91%, respectively, for one MR angiography reviewer, and of 89% and 95%, respectively, for a second reviewer for detection of stenotic disease greater than 50% when all vessel segments from the aorta through the calf vessels were considered.
For evaluation of the proximal calf just below the popliteal artery, stationary-table bolus-chase MR angiography yielded a sensitivity and specificity of 75% and 89%, respectively, for detection of stenotic disease greater than or equal to 50%. Ho et al. [9] reported sensitivity values ranging from 79% to 97% and a specificity of 100% for detection of stenotic disease greater than or equal to 50% in infrapopliteal vessels. Visualization of these smaller vessels is likely impaired by a limited contrast-to-noise ratio that may result from the combined use of a body coil for imaging and unsubtracted data for interpretation of second-station imaging. We expect that this two-station bolus-chase technique is best suited for the evaluation of ambulatory patients with claudication; in these patients, an extensive below-the-knee vascular evaluation is typically unnecessary because treatment options are focused on percutaneous therapy above the level of the trifurcation. In patients in whom depiction of below-the-knee vascular anatomy is essential (e.g., for identification of distal outflow vessels for bypass procedures or for evaluation of distal outflow vessels for planned proximal interventions in selected claudicants), a dedicated contrast-enhanced study [3, 4] or a time-of-flight study [1, 2, 5, 6] is preferable. Either of these can be performed in addition to the proposed technique.
With the exception of the external iliac artery, the results of our evaluation of proximal station vessel segments using the stationary-table method were comparable to previously reported results using conventional and moving-table MR angiography techniques [3, 4, 6, 9, 14,15,16]. In this study, a sensitivity value of 50% was obtained for the detection of greater than or equal to 50% stenosis in the external iliac artery, whereas two groups of investigators who studied conventional 3D contrast-enhanced MR angiography of the pelvic arteries reported sensitivities of 93% [14] and 100% [15] for evaluation of the external iliac artery. Sueyoshi et al. [16], another group of investigators who studied the pelvic arteries with conventional 3D contrast-enhanced MR angiography, reported a sensitivity of 100%, but they grouped common iliac vessels and external iliac vessels when deriving this value. In the moving-table literature, Ho et al. [9] reported a comparatively lower sensitivity of 89% for detection of greater than or equal to 50% stenosis of the external iliac artery.
The relatively few external iliac artery stenoses found at DSA in this study make assessment of MR angiography sensitivity calculations difficult, as supported by the wide 95% CI (0.12-0.88) that was determined for the calculated MR angiography sensitivity for this vessel segment. Nonetheless, a retrospective review of the three false-negative assessments of external iliac artery stenosis on MR angiography suggests a few additional factors that may have contributed to the low sensitivity of the stationary-table method at this level. All three vessels were found to be very close to 50% stenosis on both MR angiography and DSA. This finding is supported by the initial disagreement among MR angiography reviewers, DSA reviewers, or both, before the consensus review in all three cases.
In general, the scoring method used in this study may have contributed to reduced interobserver agreement and reduced sensitivity and specificity when the vessel segments were shown to be close to 50% stenosis on both modalities (Fig. 3A,3B,3C). A retrospective review of MR angiograms and DSA images in one patient definitively revealed that the external iliac artery appeared slightly less stenotic on MR angiography than on DSA. DSA relies on selected frontal projections for the characterization of stenotic disease, and thus it may be flawed as a standard of reference in some patients because it potentially limits the reviewer's capability to characterize the degree of lumen narrowing associated with en face atheromatous plaques (Fig. 3A,3B,3C). Thus, the use of DSA as the reference standard also creates a potential for bias against the MR angiography results.
|
|
|
This study has recognized limitations. First, the study group was derived by retrospectively identifying a group of patients who had undergone both MR angiography and DSA. This criterion resulted in a selection bias toward patients with findings positive for stenosis on MR angiography. Second, MR angiography and DSA were performed within 3 months of one another. Although each patient's chart was reviewed to ensure that no meaningful change in clinical status had been noted between the MR angiography and DSA examinations, it is possible that changes in the degree of vascular stenosis may have occurred. Such events may have contributed to discrepancies in lesions, particularly those close to 50% stenotic. Third, evaluation of four vessel segments was not possible because the corresponding DSA images were not of sufficient quality to allow characterization of these vessel segments by DSA reviewers. In our study, DSA evaluation was based on selected images that had been saved after initial dynamic imaging. It is therefore possible that subtleties in the visualization of stenotic disease best seen on dynamic imaging were missed during evaluation of the saved images. However, all vessel segments, with the exception of those discussed, were considered adequate for evaluation by both of the DSA reviewers using the available saved images. Fourth, our study group was small; to further validate our results, a larger, prospective investigation is necessary.
In conclusion, two-station stationary-table bolus-chase MR angiography provides a useful alternative to conventional catheter-based angiography. In comparison with other moving-table strategies, manual patient translation using a plastic patient-transfer board requires more manpower and coordination to successfully synchronize patient movement and imaging [9,10,11,12]. Nevertheless, in centers without moving-table technology, stationary-table MR angiography provides a simple strategy for two-station bolus-chase peripheral MR angiography.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  F. Poschenrieder, O. W. Hamer, T. Herold, T. Schleicher, I. Borisch, S. Feuerbach, and N. Zorger Diagnostic Accuracy of Intraarterial and IV MR Angiography for the Detection of Stenoses of the Infrainguinal Arteries Am. J. Roentgenol., January 1, 2009; 192(1): 117 - 121. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Ersoy and F. J. Rybicki MR Angiography of the Lower Extremities Am. J. Roentgenol., June 1, 2008; 190(6): 1675 - 1684. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Zorger, M. Volk, O. W. Hamer, M. Lenhart, J. Seitz, B. Butz, and C. Paetzel Intraarterial Gadolinium-Enhanced MR Angiography in Humans for the Detection of Infrainguinal Arterial Stenoses Before and After Percutaneous Angioplasty Am. J. Roentgenol., October 1, 2005; 185(4): 867 - 872. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
50%
