| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Radiology, Division of Interventional Radiology, University of
Virginia Health System, Box 800170, Charlottesville, VA 22908.
2 University of Virginia Medical School, Charlottesville, VA.
3 Department of Biomedical Engineering, University of Virginia, Charlottesville,
VA.
Received January 2, 2004;
accepted after revision June 3, 2004.
Address correspondence to K. D. Hagspiel.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. All experiments were performed on 1.5-T scanners. In vitro experiments were performed using 10 stents implanted into a custom-built phantom. Different orientations of the stents along the magnetic field and multiple flip angles were examined. In addition, 19 patients underwent contrast-enhanced MR angiography after the implantation of 36 stents, including four patients with six platinum stents. Angiographic correlation was available for all 19 patients, and luminal patency and stent-induced artifacts were assessed quantitatively.
RESULTS. Of the tested balloon-expandable stents, only the platinum-based stents created artifact causing luminal narrowing of 30% or less. All other balloon-expandable stents induced larger artifacts that resulted in higher degrees of narrowing. Thus, if patent, the platinum-based stents allow significant in-stent stenosis to be ruled out reliably. Selected nitinol- or tantalum-based self-expandable stents also are suitable in this regard.
CONCLUSION. Of the tested devices, platinum-based stents are the only type of currently available balloon-expandable stent that creates 30% or less artifact-induced apparent stenosis and thus are suitable for MR angiography.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The tubes were filled with a solution of 2.35 g of gadopentetate dimeglumine (Magnevist, Berlex) per liter of sterile saline (dilution, 1:200; 2.5 mmol), and MR angiography was performed. Other investigators have found that this concentration results in enhancement similar to that achieved in patients with the same sequence [3, 4]. No attempts were made to replicate the viscosity of blood. All MR angiography examinations were performed on a 1.5-T system (VISION Magnetom, Siemens Medical Solutions). A 3D fast imaging with steady-state precession (FISP) sequence was used with the following parameters: TR/TE, 4.6/1.8; bandwidth, 390 Hz/pixel; matrix, 230 x 512; and 6/8 rectangular field of view with a maximum dimension of 390 mm. The resulting voxel size was 1.54 x 0.78 x 1.5 mm.
Two sets of experiments were performed: Initially, all measurements were performed with the stents aligned perpendicular to the z-axis, and only the flip angle was changed (20°, 25°, 30°, 40°, 50°, 60°, 70°) to assess the influence of the flip angle on stent-induced artifact. In a second experiment, the measurements were performed for flip angles of 25° and 30° with the stents aligned both longitudinally and perpendicular to the main magnetic field with otherwise identical parameters to assess the influence of stent orientation on artifact generation. The phase-encoding and spatial-encoding direction was kept constant for all experiments. All examinations were performed using a head coil.
In Vivo Experiments
Patients.—Nineteen patients underwent contrast-enhanced MR
angiography after implantation of 36 stents. All patients had catheter
angiography for correlation and assessment of stent patency. The indication
for MR angiography was either evaluation of vascular territories other than
the stented segments or control of the treated segments in all cases. The time
between the MR angiography and the angiographic study varied from 1 to 330
days (mean, 71 days). In three patients, angiography was performed using
CO2 and gadolinium as contrast media because of renal
insufficiency. The other 16 angiographic examinations were performed using
iodinated contrast agents. The MR angiography and angiographic examinations
were reviewed retrospectively by two experienced vascular radiologists in
consensus. The patency of the vascular lumen on the MR angiograms was compared
with the angiographic results. The patient examinations were assessed in a
quantitative fashion using mechanical or electronic calipers. We obtained
approval from the institutional review board for this retrospective review of
patient data.
Imaging protocol.—Ten MR examinations were performed on a 1.5-T system (VISION Magnetom). A four-element phased-array coil was used for the in vivo studies. A 3D FISP sequence using the following parameters was performed: 4.6/1.8, 30° flip angle, 390 Hz/pixel bandwidth, 230 x 512 matrix, and 6/8 rectangular field of view with a maximum dimension of 390 mm. The resulting voxel size was 1.54 x 0.78 x 1.5–2.0 mm. For MR angiography, 40 mL of gadodiamide (Omniscan, Amersham Health) was injected at a rate of 1.7 mL/sec with a power injector (Spectris, Medrad). An injection delay was calculated to allow peak arterial enhancement during the acquisition of the central lines of k-space for the first acquisition. A minimum of 2 acquisitions was performed with an interscan delay of 10 sec. Nine examinations were performed on a 1.5-T high-performance gradient system (Sonata, Siemens Medical Solutions) with the following parameters: 2.5/0.98, 20° flip angle, 690 Hz/pixel bandwidth, 192 x 384 matrix, and 83.3% rectangular field of view with a maximum dimension of 400 mm. The voxel size was 1.0 x 1.7 x 1.5 mm. Three-dimensional reconstruction techniques including multiplanar reconstructions and maximum intensity projections were used. Reconstructions were performed on a standard workstation (Leonardo, Siemens Medical Solutions).
Image analysis.—Analysis of artifacts was performed in a quantitative fashion for the in vitro experiments. Lumen diameter within and outside the stent was measured from axial multiplanar reconstructions (perpendicular to the vessel median center line), and the percentage of patency of the lumen and percentage of stent artifact were calculated using vessel diameter. For the measurements, identical window level settings and electronic calipers were used. The window levels and width settings chosen were the ones that resulted in the measurements of 6 and 8 mm for the nonstented segments of the precision-drilled tubes. Measurements were rounded to the next whole number in millimeters. In cases in which the lumen was depicted in an elliptic fashion, the arithmetic mean of the two perpendicular diameter measurements was used for the calculations. The measurements were performed on a standard workstation (Leonardo, Siemens Medical Solutions).
For angiography and radiography of the stent phantom (Figs. 1A and 1B), the measurements were obtained using mechanical or electronic calipers, with the patent lumen defined as the patent area inside the stent (i.e., the stent itself was not included in the measurements). For the in vitro radiographs, 6- and 8-mm precision ball bearings were placed in the field of view for reference. The calculations were performed identical to those for MR angiography.
|
|
For the patients, two observers quantitatively assessed the visibility of the lumen on the source images, maximum intensity projections, and multiplanar reconstructions. Results were reported on a 4-point scale with 0 meaning no visibility; 1, reduced visibility, but insufficient to rule out significant stenosis (> 30% luminal narrowing); 2, reduced visibility, but sufficient to rule out significant stenosis (< 30% luminal narrowing); and 3, full visibility. Measurements were performed using mechanical or electronic calipers. In cases with observer disagreement, a consensus interpretation was performed. We chose 30% residual narrowing as the criterion for patency because it is well below the clinically accepted values for a hemodynamically significant stenosis of 50–70%.
Overall image quality, contrast bolus timing, and the presence of artifacts were assessed qualitatively on a 4-point scale for the clinical examinations. The scale for evaluating image quality was as follows: 1, excellent; 2, good; 3, poor but diagnostic; and 4, nondiagnostic. Bolus timing scores were defined as 1, excellent; 2, good; 3, fair; and 4, poor. The presence of artifacts (motion, respiratory, other) was assessed using the following scale: 1, absent; 2, present, not affecting image interpretation; 3, present, affecting image interpretation; and 4, severe. For catheter angiography, only overall image quality was assessed with the same 4-point scale as the one used for MR angiography.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The experiments with the stents aligned perpendicular to the z-axis with varying flip angles showed that of the four balloon-expandable stents tested at a 6-mm diameter, only the Omniflex stent showed less than 30% artifact-induced apparent stenosis, thus allowing assessment of the luminal patency. The other three balloon-expandable stents (Palmaz, Corinthian, and NIR) created artifacts that did not allow reliable visualization of the lumen. Both 6-mm self-expandable stents, the Wallstent and the Symphony stent, also induced less than 30% stenosis. All four stents tested at 8 mm had acceptable artifacts (Table 1). The optimal flip angles were 30° and 40° for all stents tested. Increasing the flip angle resulted in either increased artifacts or no change.
|
The second experiment, for which the flip angle was kept constant (25° and 30°) and the alignment of the stents along the main magnetic field was changed from longitudinally to perpendicular to the main magnetic field, showed improved luminal patency with the longitudinal orientation for all stents except the NIR and the Corinthian stents (Table 2).
|
In Vivo Experiments
Nineteen patients underwent contrast-enhanced MR angiography after
placement of 36 stents dilated to diameters between 6 and 9 mm.
Image quality was graded 1 (excellent) in 17 cases, as was bolus timing. In one case each, image quality was graded as good or as poor, but diagnostic. This was due to suboptimal bolus timing in both instances, ranked as good and fair, respectively. There were no motion artifacts in any of the examinations, with all 19 being graded as 1 (absent).
Twenty-seven of these 36 stents were balloon-expandable, and nine were self-expandable. There were two patients with a total of two stents in the subclavian arteries (Figs. 2 and 3), eight patients with 12 stents in the renal arteries (Figs. 4 and 5), and nine patients with 22 stents in the distal aorta and iliofemoral arteries. All stents in the subclavian and renal arteries were balloon-expandable. Of the stents placed in the distal aorta and iliofemoral arteries, 13 were balloon-expandable and nine were self-expandable stents. Six platinum stents (Vistaflex or Omniflex) were placed in four patients: one in a left subclavian artery (dilated to 7 mm), one in a right common iliac artery (dilated to 7 mm), two in two right renal arteries (dilated to 6 mm), and two in a single left renal artery (dilated to 7 mm). In all of these cases, the artifact induction was graded as 2 (reduced visibility, but sufficient to rule out significant stenosis) (Figs. 3, 5, 7A, and 7B). One patient had one Palmaz balloon-expandable stent placed in the left subclavian artery and dilated to 7 mm. Lumen visibility was rated as 0 (no visibility) (Fig. 2). Six patients had three Palmaz, three Genesis (Cordis Vascular), and two NIR (Boston Scientific) stents placed in the renal arteries and dilated to 6 mm. In all these cases, the lumen visibility was rated as 0 (Fig. 4).
|
|
|
|
|
|
Nine patients had 13 balloon-expandable stents placed in the distal aorta and iliofemoral arteries.
Besides the patient with the platinum stent, there were seven Palmaz stents (five, grade 0; two, grade 1) dilated to diameters between 7 and 9 mm. One patient had a Corinthian stent (Cordis Vascular), which is made of stainless steel, in the right common iliac artery dilated to 7 mm and an Intrastent (Sulzer Intratherapeutics), also made of stainless steel, in the contralateral common iliac artery dilated to 8 mm; neither of these stents was tested in vitro. Both stents were graded as 1 (reduced visibility, but insufficient to rule out significant stenosis) (Fig. 6).
|
One patient had a Genesis stent (Cordis Vascular) in the right common iliac artery dilated to 8 mm, graded as 1, and one patient had two Express stents (Boston Scientific) placed in the right and left common iliac arteries, dilated to 6 and 8 mm, respectively (Figs. 8A and 8B). Both were classified as grade 1.
|
|
Four patients had nine self-expandable stents placed in the iliofemoral arteries, all of which were nitinol stents. These were one Dynalink stent (Guidant) (8 mm diameter/80 mm length dilated to 6 mm), graded as 2 for visibility; two IntraCoil stents (ev3) (5/60 and 6/60 mm dilated to 5 mm) classified as grade 1; and six Protégé stents (ev3) (8/60, 7/60, 7/40 mm, all six dilated to 7 mm), also grade 1.
Table 3 shows which stents in which vascular territory allowed diagnostic MR angiography, and Table 4 gives the visibility grading for these stents.
|
|
Angiographic correlation in the 19 patients was available for all 36 stents, and vessel patency was measured as 100% within 35 stents. Image quality was graded 1 (excellent) in 16 cases; 2 (good) in one case; and 3 (poor, but diagnostic) in two cases. Both of the examinations rated as poor but diagnostic were performed with CO2 and gadolinium. One left renal artery Palmaz stent had an 80% stenosis due to intimal hyperplasia, and MR angiography showed no luminal patency (grade 0). The most characteristic feature of the platinum-based stents is their high radiodensity, which can make assessment of patency some-what difficult, especially if CO2 is used as the contrast agent (Figs. 7A and 7B).
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Interfaces of structures with significantly different magnetic susceptibilities, such as tissue and metal stents, lead to magnetic field inhomogeneities that, in turn, lead to image distortion and regional signal loss due to intravoxel dephasing. The choice of the metal used for stent construction is the primary determinant of its magnetic susceptibility and thus its MR compatibility [11]. Stents made of various stainless steel alloys show the largest artifacts, whereas materials such as nitinol and especially tantalum have been shown to offer far lower susceptibility artifacts. However, the mechanical characteristics of the two latter alloys allow the manufacture of only self-expandable stents and thus make them unsuitable for the treatment of certain lesions. Platinum is another metal with low susceptibility artifacts, but unlike the former two alloys, it is suitable for the construction of balloon-expandable stents. Copper is another suitable metal given its MR characteristics, but no commercial stent made of copper is available [16].
Loss of signal in the lumen also can result from the stent acting as a Faraday shield, which prevents penetration of radiofrequency excitation energy and blocks reception of signal from within the stent. Implants consisting of conducting loops are especially prone to the formation of these artifacts. It has been shown both theoretically and in experiments that the shielding of the protons within the cage can be overcome [14, 15, 17].
Changing the orientation of the implant in the magnetic field is one way to reduce these artifacts, but in clinical practice, this option often is not feasible. Parallel orientation of the stents in the magnetic field has been shown to increase lumen visibility [3]. Acquiring the scan with the read direction along the static magnetic field also generally results in better lumen visualization [3].
Changing the geometry of the conducting loops is another way to optimize the MR performance of stents. Different stent geometries explain the large differences in artifact size caused by stents of the same alloy (especially the ones with relatively low susceptibility) [15]. The only practical way to improve lumen visualization after implantation, however, is by increasing the radiofrequency power of the excitation pulse. This is achieved mainly by increasing the nominal flip angle. The effectiveness of this approach has been shown both in phantoms and in vivo [14, 16].
The shielding effects depend on stent materials, geometry, and orientation. One group reported that measurement of the susceptibility and shielding effects of stents is feasible, and knowledge of these effects can serve as a clinical guide when assessing the patency of stents. Large platinum stents were shown in vitro to have negligible susceptibility effects and small shielding effects [18]. Our phantom experiments confirm that platinum is indeed a metal with low susceptibility, is comparable with nitinol, and is well suited to allow luminal visualization. The new platinum-based stents allow assessment of stent patency with contrast-enhanced MR angiography both in vivo and in vitro. This improvement is an important one because balloon-expandable stents are mandatory for a variety of lesions and none of the competing currently available systems are MR-compatible in the sense of allowing assessment of stent patency. Self-expanding stents generally are more MR-compatible because they are made of either nitinol or tantalum. However, as the Protégé stents in our series show, even nitinol stents can cause significant artifacts. Unlike other authors and as predicted in theory, we did not find that increasing the flip angle improved lumen visibility within the stent. The reason for this discrepancy is probably due to differences in the scanner and sequences used for these experiments.
Our study has several shortcomings. One is that the in vivo part is retrospective rather than prospective, which would have been preferable. Also, the choice of stents for the in vitro and the in vivo experiments in this article is by no means complete; rather, it is a reflection of our institutional preferences at the time these experiments were performed, which also changed over time. In addition, we did not test all possible permutations of stent orientations within the magnetic field. The reasons for this omission are that this topic has been investigated intensively by other groups and it is not generally possible to choose the orientation once the stent has been implanted in a patient. For all practical purposes, the two orientations chosen for our experiments (longitudinal and perpendicular to the z-axis) correspond to the orientation of subclavian and renal stents and, to a lesser degree, to stents in the iliac arteries. Nevertheless, the presented data clearly show the suitability of platinum-based stents to be followed with contrast-enhanced MR angiography.
In conclusion, the platinum-based, commercially available balloon-expandable stents allow assessment of luminal patency on contrast-enhanced MR angiography using standard sequences. Stent-induced artifacts resulting in luminal narrowing do not reach a degree that would be considered hemodynamically significant.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  M. B. Blum, M. Schmook, R. Schernthaner, G. Edelhauser, S. Puchner, J. Lammer, and M. A. Funovics Quantification and Detectability of In-Stent Stenosis with CT Angiography and MR Angiography in Arterial Stents In Vitro Am. J. Roentgenol., November 1, 2007; 189(5): 1238 - 1242. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M.-C. P. Shih, J. F. Angle, D. A. Leung, K. J. Cherry, N. L. Harthun, A. H. Matsumoto, and K. D. Hagspiel CTA and MRA in Mesenteric Ischemia: Part 2, Normal Findings and Complications After Surgical and Endovascular Treatment Am. J. Roentgenol., February 1, 2007; 188(2): 462 - 471. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O W Hamer, I Borisch, C Paetzel, W R Nitz, J Seitz, S Feuerbach, and N Zorger In vitro evaluation of stent patency and in-stent stenoses in 10 metallic stents using MR angiography Br. J. Radiol., August 1, 2006; 79(944): 636 - 643. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
