| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
Institute of Diagnostic Radiology, University Hospital Zurich, CH-8091 Zurich,
Switzerland.
2
Present address: Department of Diagnostic Radiology, University Hospital
Essen, Hufelandstr. 55, D-45122, Essen, Germany.
3
Institute of Angiology, University Hospital Zurich, CH-8091 Zurich,
Switzerland.
Received March 26, 1999;
accepted after revision September 23, 1999.
Address correspondence to J. F. Debatin.
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Both conventional digital subtraction angiography and three-dimensional contrast-enhanced MR angiography with a dedicated peripheral vascular coil were performed in 61 patients with suspected peripheral vascular disease. In a prospective analysis, one reviewer evaluated the digital subtraction angiographic images and a second reviewer evaluated the MR angiographic images; both were unaware of the results of the other imaging technique. Each vascular segment (29 segments per patient) was evaluated for the presence of occlusive vessel disease. The following grading system was applied: 0, normal; 1, vessel irregularity with a luminal reduction of less than 10%; 2, mild stenosis (lumen reduction, 10-49%); 3, severe stenosis (lumen reduction, 50-99%); and 4, occlusion (lumen reduction, 100%). In 11 patients surgical graft patency was assessed.
RESULTS. MR angiography provided an image quality comparable with that of digital subtraction angiography. Overall sensitivity and specificity for MR angiography were 92% and 96.6%, respectively, for the detection of hemodynamically significant disease and 92.3% and 99.4%, respectively, for the detection of occlusions.
CONCLUSION. Two-station contrast-enhanced three-dimensional MR angiography with a dedicated lower-extremity vascular coil proved effective enough to consider it as a noninvasive alternative to digital subtraction angiography in the assessment of the pelvic and lower extremity arterial vasculature.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
In recent years, we have seen the development and rapid clinical implementation of contrast-enhanced three-dimensional MR angiography for the assessment of virtually all vascular territories [9,10,11,12,13,14]. This technique is based on the availability of high-performance gradient systems, capable of reducing data collection times sufficiently to acquire a three-dimensional image set during the intravascular phase of an IV-administered extracellular contrast agent. Subsequent parenchymal enhancement and contrast dose limitations initially curtailed use of the technique to the display of the vascular territory contained within a single 40- to 48-cm field of view. Concomitant assessment of the pelvic and runoff arteries, however, requires extended coverage of the arterial system from the aortic bifurcation to the distal runoff vessels.
The purpose of this study was to determine the diagnostic accuracy of a three-dimensional MR angiographic strategy, one that combined use of a dedicated vascular coil with performance of a single-injection, two-station protocol, with regard to the assessment of the arterial vasculature from the aortic bifurcation to the distal trifurcation vessels in patients with peripheral vascular disease. Conventional DSA served as the standard of reference.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
MR Imaging
All MR imaging was performed on a 1.5-T system (Signa EchoSpeed; General
Electric Medical Systems, Milwaukee, WI) equipped with a high-performance
gradient system. Patients were wrapped in a multichannel quadrature-phased
array peripheral vascular coil (Peripheral Vascular Array; Medical Advances,
Milwaukee, WI) and were imaged in a supine position. The coil consists of four
circular arrays and is used for signal reception. The flexible design of the
coil allows bilateral vascular imaging in close proximity to the anatomy of
interest resulting in high spatial and contrast resolution. Each element
covers a territory of 24 cm (total coverage, 96 cm) and can be activated
separately or in combination with one other element. The coil was placed to
encompass the pelvic and runoff vasculature from the aortic bifurcation to the
distal trifurcation vessels (Fig.
1). The first two coil elements covered the pelvis and thighs, and
the second set covered the popliteal and trifurcation vessels.
|
The imaging strategy was based on the acquisition of two three-dimensional data sets, each time using two adjacent coil elements extending more than 48 cm. To determine the anteroposterior offset of the two three-dimensional acquisition volumes, 12 10-mm axial two-dimensional time-of-flight images were acquired at 10-cm intervals through the pelvis, thighs, and lower limbs. Based on these images, the two three-dimensional acquisitions, each consisting of 48 contiguous sections, were prescribed. Section thickness was individually adapted to ensure coverage of the entire vascular territory and ranged between 2.4 and 2.8 mm. The applied three-dimensional spoiled gradient-recalled echo sequence used the following parameters: TR/TE, 5.2/1.5; inversion time, 28 msec; and flip angle, 30°. A 48 x 36 x 12 cm field of view combined with a 256 x 192 x 48 matrix size rendered an in-plane spatial resolution of 1.8 x 1.8 x 2.4 mm. An interpolation scheme was applied (zero interpolation) with adding of extra zeroes to the K-space data in all three planes before the Fourier transformation. Thus, the pixel size was halved resulting in an improvement of the apparent resolution to 0.9 x 0.9 x 1.2 mm. With this technique, partial volume effects can be reduced by making the pixel size smaller. Each image set was collected over 30 sec: the first covered the pelvis and thighs, and the second covered the popliteal and trifurcation vessels.
Before collecting the three-dimensional data sets, the scan delay for the first three-dimensional acquisition after beginning administration of contrast material into the antecubital vein was determined. For this purpose, axial multiphase gradient-echo images were collected at the level of the lower thigh after administration of a 2-ml contrast test bolus. The contrast agent (gadopentetate dimeglumine [Magnevist]; Schering, Berlin, Germany) was administered at a rate of 0.6 ml/sec by an automated injector (MR Spectris; Medrad, Pittsburgh, PA) through a 20-gauge needle placed in the antecubital fossa and was followed by a 15-ml normal saline flush. A signal-intensity curve was determined for a region of interest placed within the superficial femoral artery on both sides. Half time to maximum signal intensity was determined in the delayed leg and was used as the scan delay (Fig. 2).
|
Both three-dimensional data sets were collected over 70 sec: 30 sec for each of the two three-dimensional acquisitions separated by a 10-sec imaging break. During the 10-sec break, the MR table was manually repositioned to the center of the lower imaging volume (Fig. 3A,3B,3C), which was offset by only 45 cm to ensure some overlap between the two data sets. Craniocaudal coverage thus extended over 90 cm. Gadopentetate dimeglumine was administered over 70 sec followed by a saline flush of 20 ml using the automated injector. Based on patient size, the flow rate was adapted to the volume that corresponded to a dose of 0.3 mmol/kg, ranging between 0.5 and 0.7 ml/sec.
|
|
|
MR angiographic data sets were postprocessed on an Advantage Windows workstation (General Electric Medical Systems). Maximum intensity projections were rendered containing the proximal (distal aorta and pelvic and upper thigh arteries) and distal (popliteal and trifurcation arteries) vascular territories. Rotated maximum-intensity-projection displays ranging from -60° to +60° were documented on film. In addition, images of each territory were merged in the multiple display mode to provide an overview of the entire runoff.
Conventional DSA
DSA of the pelvic and lower extremity vessels was performed on two standard
angiography units (Digitron 3, Siemens, Erlangen, Germany; or Integris V3000
[version 13], Philips Medical Systems, Best, the Netherlands). All 61 patients
underwent catheter angiography extending from the distal aorta to the lower
trifurcation vessels with a transfemorally inserted 5-French pigtail catheter
(angiographic catheter; AngioDynamics, Queensbury, NY). The catheter tip was
positioned just above the aortic bifurcation for DSA of the pelvic arteries,
followed by multiple acquisitions encompassing the thigh and lower limbs. At
each station, 20 ml of iodinated contrast material (ioxaglate [320 mg/ml]) was
administered. As required, the examination was supplemented by acquisition of
one or more oblique views of the pelvic arteries using 15 ml of contrast
material.
Image Analysis
DSA and MR angiographic images were interpreted prospectively. DSA images
were interpreted by a board-certified radiologist specializing in vascular
interventions who was unaware of the MR angiographic data, and MR angiographic
images were reviewed by a board-certified radiologist with special training in
MR angiography who was unaware of the DSA results. Analysis of both
examinations was based on all images. DSA images and the
maximum-intensity-projection displays were documented on film using similar
magnification factors and could be analyzed on a monitor for further
interpretation. Three-dimensional MR angiographic data sets were available on
a workstation permitting review of the source images as well as interactive
reformation at the time of interpretation. DSA was used as the standard of
reference.
For analysis, the arterial system was divided into the following segments: 1, distal infrarenal aorta; 2, common iliac artery; 3, internal iliac artery; 4, external iliac artery; 5, common femoral artery; 6 and 7, superficial femoral artery divided into proximal and distal halves; 8, popliteal artery; 9, tibiofibular trunc; 10 and 11, anterior tibial artery divided into proximal and distal segments; 12 and 13, peroneal artery, divided into proximal and distal segments; and 14 and 15, posterior tibial artery divided into proximal and distal segments.
Segments not contained within the imaging volume were considered nondiagnostic. All other segments were evaluated for the presence of stenotic or aneurysmal disease. Occlusive disease was graded on a 4-point scale on the basis of the most severe reduction of the arterial diameter compared with the most normal-appearing segment proximal or distal to the area of arterial compromise: 0, normal vessel; 1, irregularity of the vessel wall caused by atherosclerotic plaque formation with less than 10% luminal reduction; 2, stenosis with less than 50% luminal narrowing; 3, stenosis with 50% luminal narrowing or more; and 4, occlusion.
Sensitivities and specificities were calculated for all segments together and for each vessel separately. When two or more pathologic changes were present in one vessel segment, the most severe change was used for subsequent grading and analysis. Kappa values were calculated to measure concordance between the results of DSA and MR angiography.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
|
|
|
|
|
|
|
MR angiographic images failed to adequately depict 58 segments (3%). In five patients the superior acquisition volume was centered too low, resulting in the failure to depict the distal aorta (n = 5) and common iliac artery (n = 6). Technical problems impaired assessment of MR angiographic images in two other patients. A failure to reconstruct the distal data set for an unknown reason rendered 18 segments nondiagnostic. In the second patient, detection of the test bolus was erroneously based on signal changes within a small collateral vessel that had filled in a retrograde fashion. The scan delay was miscalculated and the bolus passage was missed resulting in the inability to evaluate the entire examination (29 segments). Subsequent analysis was thus based on 1711 segments. MR angiographic images showed 723 segments as normal, 399 segments as irregular with less than 10% luminal narrowing, 224 segments as mildly stenosed, and 168 segments as severely stenosed. One hundred ninety-seven segments were considered occluded (Table 1).
Table 2 summarizes the MR angiographic results relative to the conventional angiographic findings. Severe stenosis exceeding 50% luminal narrowing was identified on both DSA and MR angiography in 126 segments. Thirty-nine segments were overgraded as severely stenosed using MR angiography, whereas using DSA these segments were characterized as normal (n = 5), irregular (n = 8), or mildly stenosed (n = 26) (Figs. 7A,7B,7C and 8A,8B,8C,8D). Using MR angiography, underestimation of severe stenoses as either normal (n = 1), irregular (n = 2), or mild (n = 12) occurred in 15 segments (Fig. 9A,9B). Using both techniques, 186 segments were characterized as occluded. Based on MR angiographic images, reviewers overgraded eight segments as occluded that were characterized as severely stenosed using DSA. Using MR angiography, 17 segments of distal runoff vessels that had been interpreted as occluded on DSA were found to be patent, containing no abnormality (n = 1), irregularity (n = 8), or mild (n = 5) or severe (n = 3) stenosis.
|
|
|
|
|
|
|
|
|
|
Combining severe stenoses and occlusions, the overall sensitivity and specificity for detection of occlusive disease was 92.2% and 98%, respectively; sensitivity and specificity values were 92.3% and 99.4%, respectively, when only occluded segments were considered. For a 95% confidence interval ranging from 0.83 to 0.87, the kappa value calculated to measure concordance between DSA and MR angiography was 0.85 for the detection of mild stenoses, severe stenoses, and occlusions.
Aneurysmal changes were identified as such on both DSA and MR angiography in nine segments. Ten femoropopliteal grafts and one vascular graft extending from the common femoral artery to proximal anterior tibial artery were found to be patent on both DSA and MR angiographic images. Artifacts caused by metallic clips were readily identified on the individual source images of the MR angiographic data sets because of characteristic signal dropout in the region of the clip (Fig. 10A,10B).
|
|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The lower extremity arterial system has been targeted by a variety of MR angiographic strategies. Based on the acquisition of thin axial slices, two-dimensional time-of-flight techniques were found to be highly sensitive to slow flow [15,16,17,18,19,20,21,22], providing results superior to even conventional arteriography for identification of distal runoff vessels [23]. The limited clinical impact of the technique mainly reflects the lengthy acquisition time, often extending beyond 30 min for a single extremity [9, 24], and magnet time, taking up to 90 min, needed to assess the vasculature from the aorta to the lower limb arteries [25]. Because contrast-enhanced three-dimensional MR angiography does not rely on time-of-flight effects it permits imaging along the major-vessel axis [26], thereby vastly shortening imaging times to a total magnet time of less than 25 min for the display of the vasculature from the distal aorta to the trifurcation vessels. Although contrast-enhanced three-dimensional MR angiography has been shown to perform well in the evaluation of the pelvic arteries [10, 14, 27], the apparent inability to cover the entire arterial tree from the aortic bifurcation to the lower extremity runoff vessels has long been considered a limitation.
Based on the observation that the capillary passage of extracellular contrast material sufficiently reduces the concentration of venous contrast material for select display of the lower extremity arterial system, Ho et al. [28] and Meaney et al. [25] recently advocated a slow-infusion, multistation technique. Paramagnetic contrast material was infused over several minutes while three-dimensional data were collected at successive stations. The two-station imaging strategy we used in this study is based on the same principle. The use of a dedicated surface coil with very short TRs provides an excellent contrast-to-noise ratio, obviating image subtraction. A short data collection period extending over a mere 70 sec permits injection of contrast material at high rates, resulting in good delineation of even small trifurcation vessels.
The two-station, single-bolus—injection protocol proved highly robust, with few technique-related problems. Although the failure to reconstruct one distal image set remains unexplained, the analysis of a miscalculated timing regimen points to a possible pitfall. In a patient with extensive arterioocclusive disease, the detection of the test bolus was inadvertently performed in a collateral vessel, which apparently had filled in a retrograde fashion. The bolus passage was missed, leaving insufficient signal in the arterial system to render diagnostic images. Retrospectively, the detection of the test bolus should have been repeated at a different level. This case points to the dependence of image quality on proper timing of the IV contrast bolus. For optimal results, the arterial concentration of contrast material should plateau while image data are collected [29]. To define the travel time from the peripheral injection site to the runoff vessels, a test bolus should be used [30]. To ensure the presence of contrast material in all relevant vessels, timing should be based on measurements of the delayed leg.
The presented strategy provided diagnostic image quality in all patients. Some venous enhancement in the distal volume was found in patients with high-grade chronic disease affecting proximal vessel segments. This observation suggests that the presence of extensive collaterals results in a faster passage of contrast material into the venous system. In all patients, venous overlap was easily compensated for by viewing the data set in the interactive multiplanar reformation mode.
Good image quality is reflected by the comparative analysis with DSA. Severe stenoses were correctly identified on MR angiography with an overall sensitivity and specificity of 92% and 96.6%, respectively. Overestimation of stenoses occurred more frequently (n = 39) (Figs. 7A,7B,7C and 8A,8B,8C,8D) than underestimation (n = 15) (Fig. 9A,9B). Correlation between MR angiography and DSA was lower in cases of mild disease (Table 2), but even in these cases sensitivity and specificity values ranged between 70% and 90%. The technique also proved adequate for assessment of the morphology of surgically reconstructed arterial segments. Thus, graft patency was correctly confirmed in 11 patients. Signal voids caused by metallic clips, which may simulate the presence of a severe stenoses, need to be considered (Fig. 10A,10B).
For lack of a more accurate technique, we used DSA as a reference standard.
Limitations of this projectional technique for the detection and
characterization of nonconcentric stenoses are well recognized. Seventeen
runoff vessel segments, interpreted as occluded on DSA but found patent on
three-dimensional MR angiography, underscore these concerns. Similar
observations have been made in other vascular territories such as the renal
arteries [31]. Another
potential cause for differences between MR angiography— and DSA-based
interpretations relates to the presence of interobserver variability (Fig.
9A,9B).
Thus, the kappa value between the MR angiography and DSA interpretations (95%
confidence interval, 
= 0.85 ± 0.02) is similar to that reported
for observers viewing the same conventional DSA image sets
[32].
The presented strategy does have several drawbacks. First and most important, craniocaudal coverage remains limited. The overlap between the two acquisitions reduces coverage to 90 cm. This precludes inclusion of the renal arteries in the imaging volume in adult patients. In tall patients, inclusion of the distal aorta proved difficult, resulting in incomplete coverage of the vascular tree in five of the 61 patients. These five patients had an average height of 186 cm, ranging from 183 to 191 cm. Because magnet hardware limits the field of view to 48 cm, this issue can only be resolved by adding a third acquisition, similar to the technique proposed by Meaney et al. [25]. Coverage would then extend more than 130 cm—sufficient for the additional display of the renal arteries. Finally, the current strategy would benefit from automation of table motion. Software solutions have already been implemented in some scanners.
This study should be considered another step toward clinical implementation of MR angiography for assessment of the peripheral arterial vasculature. The presented two-step, single-injection strategy is robust and accurate in the evaluation of peripheral vascular disease. Eminent hard- and software improvements including the design of longer coils and automation of table motion promise to add to the diagnostic performance of this technique in the foreseeable future.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  D. R. Hadizadeh, J. Gieseke, S. H. Lohmaier, K. Wilhelm, J. Boschewitz, F. Verrel, H. H. Schild, and W. A. Willinek Peripheral MR Angiography with Blood Pool Contrast Agent: Prospective Intraindividual Comparative Study of High-Spatial-Resolution Steady-State MR Angiography versus Standard-Resolution First-Pass MR Angiography and DSA Radiology, September 3, 2008; (2008) 2492072033. [Abstract] [Full Text]
|
|
|
|
 |
|
 K. Nael, S. G. Ruehm, H. J. Michaely, R. Saleh, M. Lee, G. Laub, and J. P. Finn Multistation Whole-Body High-Spatial-Resolution MR Angiography Using a 32-Channel MR System Am. J. Roentgenol., February 1, 2007; 188(2): 529 - 539. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. M. Kelly, P. Cronin, H. K. Hussain, F. J. Londy, D. B. Chepeha, and R. C. Carlos Preoperative MR Angiography in Free Fibula Flap Transfer for Head and Neck Cancer: Clinical Application and Influence on Surgical Decision Making Am. J. Roentgenol., January 1, 2007; 188(1): 268 - 274. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Andreisek, T. Pfammatter, K. Goepfert, D. Nanz, P. Hervo, R. Koppensteiner, and D. Weishaupt Peripheral Arteries in Diabetic Patients: Standard Bolus-Chase and Time-resolved MR Angiography Radiology, December 19, 2006; (2006) 2422051111. [Abstract] [Full Text]
|
|
|
|
|
|
F. S. Pereles, J. D. Collins, J. C. Carr, C. Francois, M. D. Morasch, R. M. McCarthy, E. A. Krupinski, G. M. Butler, and J. P. Finn Accuracy of Stepping-Table Lower Extremity MR Angiography with Dual-Level Bolus Timing and Separate Calf Acquisition: Hybrid Peripheral MR Angiography Radiology, July 1, 2006; 240(1): 283 - 290. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Tongdee, V. R. Narra, G. McNeal, C. F. Hildebolt, F. El-Merhi, G. Foster, and J. J. Brown Hybrid peripheral 3D contrast-enhanced MR angiography of calf and foot vasculature. Am. J. Roentgenol., June 1, 2006; 186(6): 1746 - 1753. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. J. Schaefer, F. P. Boudghene, H. J. Brambs, M. Bret-Zurita, J. L. Caniego, R. A. Coulden, H. B. Gehl, R. Hammerstingl, A. Huber, R. J. Mendez, et al. Abdominal and Iliac Arterial Stenoses: Comparative Double-blinded Randomized Study of Diagnostic Accuracy of 3D MR Angiography with Gadodiamide or Gadopentetate Dimeglumine Radiology, March 1, 2006; 238(3): 827 - 840. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Fenchel, A. M. Scheule, N. I. Stauder, U. Kramer, K. Tomaschko, T. Nagele, C. Bretschneider, H.-P. Schlemmer, C. D. Claussen, and S. Miller Atherosclerotic Disease: Whole-Body Cardiovascular Imaging with MR System with 32 Receiver Channels and Total-Body Surface Coil Technology--Initial Clinical Results Radiology, December 1, 2005; 238(1): 280 - 291. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. K. Willmann, B. Baumert, T. Schertler, S. Wildermuth, T. Pfammatter, F. R. Verdun, B. Seifert, B. Marincek, and T. Bohm Aortoiliac and Lower Extremity Arteries Assessed with 16-De |
