BJR
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
First published online April 26, 2006
British Journal of Radiology (2006) 79, 636-643
© 2006 British Institute of Radiology
doi: 10.1259/bjr/57301879
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hamer, O W
Right arrow Articles by Zorger, N
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hamer, O W
Right arrow Articles by Zorger, N

Full paper

In vitro evaluation of stent patency and in-stent stenoses in 10 metallic stents using MR angiography

O W Hamer, MD I Borisch, MD C Paetzel, MD W R Nitz, PhD J Seitz, MD S Feuerbach, MD and N Zorger, MD

Department of Radiology, University Hospital of Regensburg, Franz-Josef-Strauss-Allee 11, 93042 Regensburg, Germany


   Abstract
Top
 Abstract
Introduction
 Materials and methods
 Results
 Discussion
 References
 
In vitro study to investigate the suitability of contrast enhanced magnetic resonance angiography (CEMRA) for determination of stent patency and grading of in-stent stenoses in 10 metallic stents. The Acculink carotid, DynaLink, Easy Wallstent, JostentSelfX XF, Luminexx, Omnilink, sinus-SuperFlex, SMART, Symphony and ZA stent were separately placed in a vascular phantom. Dedicated stenoses inside the stents generated a concentric lumen narrowing of 50%. CEMRA was performed for each stent. Signal loss inside the stents and artificial lumen narrowing were assessed objectively using the evaluation software of the MR imager. Moreover, three blinded observers determined visibility of stent patency and in-stent stenoses subjectively on a 3-point scale and graded in-stent stenoses. Loss of signal intensity within the stent lumen ranged between 90% (Wallstent) and 5% (ZA), artificial lumen narrowing between 56% (Symphony) and 22% (ZA). For the Symphony and Wallstent, visibility of patency and in-stent stenoses was impaired and the observers' grading exaggerated the degree of stenoses (by 23% and 33%, respectively). For the remainder of stents, patency and stenoses were visible and stenoses were graded accurately (less than 10% discrepancy from reference standard). In this in vitro study, eight of 10 stents presented with MRI characteristics which enabled determination of stent patency and accurate grading of clinically relevant in-stent stenoses.


   Introduction
Top
 Abstract
 Introduction
Materials and methods
 Results
 Discussion
 References
 
Contrast-enhanced three-dimensional magnetic resonance angiography (CEMRA) is a non-invasive alternative to digital subtraction angiography (DSA) and has meanwhile gained wide acceptance regarding the assessment of arterial occlusive disease in virtually all arterial territories [1–5]. Patient monitoring after stent placement is still a challenge for CEMRA due to image distortion by artefacts emanating from the stent. Computed tomography angiography (CTA) has been advocated as a suitable technique for follow-up after stent implantation. However, CTA is associated with the need for ionizing radiation and potentially harmful iodinated contrast material [6, 7]. Moreover, it has been demonstrated that stent visualization on CTA images can be hampered by markedly artificial lumen narrowing and pseudoenhancement within the stent lumen [8]. Colour duplex ultrasound is another non-invasive technique which is capable of assessing stent patency and eventual in-stent stenoses. Performance of this technique, however, can be very difficult in patients with obesity and gaseous distension of the bowel when iliac stents are the target. However, monitoring of endoprostheses in iliac arteries is of particular interest because iliac stenting is a frequently performed procedure.

Several experimental studies have been performed to evaluate the nature and amount of artefacts of a variety of stent designs and materials on MR images [9–16]. Some of these studies evaluated whether stent patency can be determined, reporting promising results for a subgroup of stents [17–24]. However, for a clinically valuable follow-up technique, it is crucial to detect not only stent patency, but also significant in-stent stenoses.

The purpose of this in vitro study simulating iliac anatomy was to investigate the suitability of CEMRA for determining stent patency as well as detecting clinically relevant in-stent stenoses in 10 metallic stents.


   Materials and methods
Top
 Abstract
 Introduction
 Materials and methods
Results
 Discussion
 References
 
Pre-study to determine the MR background signal
MR background signal in vivo
Within a pre-study, the MR signal of perivascular pelvic soft tissue was retrospectively determined for 27 patients who had received a CEMRA of the iliac vasculature for clinical purposes. The CEMRA sequence was identical to the sequence used in the main in vitro study (see below). The coronal source image best showing the aortic bifurcation was determined for each patient. Using the MR evaluation software (NUMARIS 3.5 a1.1b; Siemens AG, Medical Solutions, Erlangen, Germany), one of the authors placed a circular user-defined region of interest (ROI) with a diameter of 3–4 cm in the soft tissues beneath the aortic bifurcation of each patient. The mean signal intensity as well as standard deviation (SD) was registered for each patient. The mean signal intensity±SD for all 27 patients was calculated.

MR background signal in vitro
Blankets which had been soaked in water at room temperature were positioned in the centre of the magnet and imaged with the identical CEMRA sequence as applied in the main study (see below). The same author who had performed the in vivo measurements for the 27 patients placed ROIs with a diameter of 3–4 cm on a coronal source image of the blankets and determined the mean signal intensity±SD.

Main in vitro study to determine the MRI characteristics of 10 metallic stents
Stents
Ten metallic stents were evaluated (Acculink carotid, DynaLink, JostentSelfX XF, Luminexx vascular, Omnilink, sinus-SuperFlex, SMART, Symphony, Easy Wallstent, and ZA-stent). The dimensions, material composition and manufacturers are listed in Table 1Go. Nine of these 10 stents were self-expanding and one was balloon-expanding (Omnilink). All stents had a dedicated diameter of 10 mm and thus were suitable for the treatment of iliac stenoses (the Acculink carotid is intended for the treatment of carotid stenoses; the stent's dimensions, however, render an implantation in the iliac vasculature possible also). The stents' lengths ranged from 38 mm to 60 mm (mean 45.0 mm±7.6 mm).


View this table:
[in this window]
[in a new window]
  		Table 1. Technical stent parameters and manufacturers

 
Vascular phantom
The stents were placed in a vascular phantom (endovascular model # 1803; Sawbones, Malmö, Sweden). The phantom consisted of tubes made from flexible urethane polymer and simulated the abdominal aorta, renal arteries, common iliac arteries and external iliac arteries of both sides. The common and external iliac arteries were made out of the same tube and thus were of identical dimensions without noticeable transition. All stents were separately implanted in the right common/external iliac artery of the phantom, which had an inner diameter of 10 mm and a wall thickness of 2 mm. Because the tubes were transparent, complete stent opening and attachment to the tube wall was assured visually. The phantom was filled with a 2.00 mmol l–1 gadopentetate dimeglumine in saline solution at room temperature: 1 l of saline solution was mixed with 4 ml of gadopentetate dimeglumine (0.5 mol l–1; Magnevist; Schering GmbH, Berlin, Germany). This dilution was chosen according to settings used by other authors who applied similar MR sequences [11, 13, 25]. The MR signal value of this dilution had been shown to be similar to the signal of contrast filled vessels on CEMRA images in vivo. The phantom was embedded in water soaked blankets to simulate perivascular pelvic soft tissue and to enable proper shimming.

Artificial in-stent stenosis
A defined lumen narrowing was created by placing a dedicated stenosis made from rigid urethane (model # 1802-2; Sawbones, Malmö, Sweden) in each stent. The artificial stenoses were 20 mm in length and cylindrically shaped, showing a central lumen and thus simulating concentric stenoses. The lumen narrowing for all employed stenoses was set to be 50%.

MRA imaging
All images were acquired at a commercial 1.5 T MRI unit (Magnetom Sonata; Siemens AG, Medical Solutions, Erlangen, Germany). The vascular phantom was positioned in the centre of the magnet. The amount of susceptibility artefacts caused by the stent depends on the orientation of the stent compared with the main magnetic field B0. The distortion of the magnetic field by a cylindrical object is minimal when the cylinder's long axis is aligned with the direction of the main magnetic field [9, 12, 14]. The artefacts grow when the object is rotated toward a direction in which its long axis is perpendicular to B0. This dynamic was demonstrated in several experimental studies for numerous stents [13, 15]. With the purpose of comparing the stent specific artefacts rather than the orientational dependence, it was decided to choose one orientation of the stent containing tube to B0, which is representative for the common/external iliac arteries. The image distortion by all of the stents would be more severe for an angle setting more perpendicular to B0 and less severe for an angulation more parallel to B0. However, the relative severity of image distortion and thus the ranking regarding suitability for CEMRA imaging seen for the 10 evaluated stents would be identical, independent of the specific angle. Thus, in order to identify those stents which are best suitable for CEMRA and which should be chosen if follow-up with CEMRA is intended, one angle setting is sufficient as long as the setting is identical for all stents. Therefore, the tube which contained the stent was orientated at a lateral deviation of 30° as well as an upward deviation of 45° to the z-axis and main magnetic field (B0), respectively, to simulate the course of the common/external iliac artery. A phased-array body coil (circularly polarized) served as a receiver coil. Oblique coronal source images (parallel to the orientation of the stent containing tube) were obtained. The applied CEMRA sequence was a three-dimensional gradient-echo sequence with RF spoiling (fast low-angle shot, repetition time (TR) 3.37 ms, echo time (TE) 1.24 ms, flip angle 25°, matrix 246 x 512, read-out bandwidth 390 Hz pixel–1, 1.1 mm partition thickness (after Fourier interpolation), voxel size 1.3 mm x 0.7 mm x 1.1 mm, frequency encoding parallel to the main magnetic field) identical to the sequence usually used for CEMRA of the pelvis in the authors' department. For each stent, one angiographic data set was acquired. The data sets were post-processed by application of a standard maximum intensity projection (MIP) algorithm (22.5° rotational intervals around the craniocaudal axis covering 180°). Finally, the images were sent to a MagicView workstation (Siemens AG).

Subjective data analysis
Three radiologists, each with 4 years of experience in MR angiogram interpretation, evaluated the images independently on softcopy displays at a MagicView workstation. The observers were blinded regarding the type of stent and degree of in-stent stenosis. Each observer evaluated each stent once; there were no repeated measurements. They were instructed to rank their overall impression of the quality of each CEMRA study (source and MIP images were evaluated together for this analysis) on a 4-point scale (1 = good, 2 = minor limitations, 3 = major limitations, 4 = not diagnostic). Furthermore, both stent patency and visibility of in-stent stenoses had to be ranked on a 3-point scale (1 = visible, 2 = limited visibility, 3 = not visible). This was done separately for source and MIP images. If "limited visibility" or "visible", the degree of stenosis had to be determined at its narrowest location on the source images. The reference diameter was defined as the diameter of the tube segment adjacent to the proximal stent end. The severity of stenosis had to be expressed as a percentage of the reference diameter. The mean±SD of the three observers' gradings and the difference of this mean to 50% (reference standard) were computed for every stent.

Objective data analysis
For each stent, the author who had performed the measurements of the pre-study assessed (1) the background signal intensity on CEMRA images, (2) the apparent diameter of the stent lumen outside the artificial stenosis and (3) the loss of signal intensity within the stent outside the artificial stenosis. The author was blinded to the type of stent and acquired one data set per stent. Measurements were made on source images by using the evaluation software of the MR imager (NUMARIS 3.5 a1.1b; Siemens AG, Medical Solutions, Erlangen, Germany).

Ad (1): The background signal intensity was determined as follows: ROIs with a diameter of 3–4 cm were placed beneath the "aortic bifurcation" of the vascular phantom on coronal source images. The mean signal intensity±SD was determined for each of the 10 studies. The mean signal value±SD of these 10 signal measurements was calculated.

Ad (2): The apparent stent lumen and artificial lumen narrowing, respectively, were determined as follows: signal intensity plots were drawn perpendicular to the long axis of the stent. The plots were placed within the visible stent lumen outside the in-stent stenosis. The location of the artificial in-stent stenosis was easily visible for eight stents (Acculink, DynaLink, JostentSelfX XF, Luminexx, Omnilink, sinus-SuperFlex, SMART and ZA). For two stents (Symphony and Wallstent), the margins of the in-stent stenoses could not be determined unequivocally on MR images. However, the examiner was aware of the exact location because he knew the experimental setup. The outer contour of the visible stent lumen was determined using a method similar to that used by other groups [13, 26]: Those two points (left and right margin of the stent lumen) on the signal intensity plot were identified where the signal exceeded twice the signal of the background. The apparent stent lumen diameter was defined as the distance between these two points (Figure 1Go). The ratio of apparent stent lumen diameter and the respective value of the reference tube, which was the tube segment adjacent to the proximal stent end, was calculated and represented the artificial lumen narrowing (the higher the value, the less artificial lumen narrowing, and vice versa).


Figure 1
View larger version (10K):
[in this window]
[in a new window]
  		Figure 1. Schematic of signal intensity plot(dotted line) perpendicular to long axis of stent. The apparent stent lumen was defined to be the distance between the two points on the plot where the signal exceeded twice the background signal.

 
Ad (3): The signal intensity within the stent was determined as follows: a circular ROI was placed (a) within the visible stent lumen without covering parts of the stent struts or the in-stent stenosis and (b) within the reference tube. The mean signal values within both ROIs were measured. The loss of signal intensity within the stent was determined by calculating the ratio of the signal within the stent and the signal in the reference tube for each stent (the higher the value, the less signal loss, and vice versa).

To compare the suitability of the stents for CEMRA, three groups were established on the basis of the degree of artificial lumen narrowing and loss of in-stent signal intensity. The thresholds were chosen according to the strategy published by other authors [13]. Group 1 stents were stents with a minor reduction of signal intensity within the stent and a minor artificial lumen narrowing (signal intensity and apparent stent lumen 66–100% relative to the reference tube). These stents were considered to be well suited for CEMRA. Group 2 stents were considered to be partially suited for CEMRA: one of the two parameters (signal intensity or apparent stent lumen) was less than 66%. Group 3 stents were considered to be not well suited for CEMRA: both parameters were less than 66%.

Statistics
Interobserver variability for categorical data (ranking of visibility stent patency and in-stent stenoses) was determined using kappa statistics. In case kappa statistics was not applicable because the cross tabulated scores were not symmetrical, Spearman's correlation coefficient was calculated. For continuous data (severity of in-stent stenoses), kappa statistics was not applicable because calculation of kappa requires that all observers use the same rating categories. In the present study, however, the observers' grading of stenoses was not classified, but resulted in a large number of possible responses. Thus, interobserver variability regarding determination of severity of in-stent stenoses was evaluated by calculating Pearson's correlation coefficient. A p-value ≤ 0.05 was considered to be statistically significant.


   Results
Top
 Abstract
 Introduction
 Materials and methods
 Results
Discussion
 References
 
MR background signal
The mean background signal intensity, as retrospectively assessed for the 27 in vivo studies, was 10.4±3.9. The background signal intensity determined within the pre-study for the water soaked blankets was 13.3±3.5. Because the difference between these two signal values was negligible, the water soaked blankets were considered to be suitable to simulate perivascular pelvic soft tissue, and thus were applied in the main study. The mean background signal intensity for the 10 experimental stent studies was 14.8±3.9.

Subjective data analysis
The source and MIP images of all 10 stents are given side by side in Figure 2Go.


Figure 2
View larger version (83K):
[in this window]
[in a new window]
  		Figure 2. Source images(upper row) and maximum intensity projection (MIP) images (lower row) obtained from oblique coronal 3D contrast enhanced MR angiography (CEMRA) images of 10 metallic stents. The stents from left to right: group 1: DynaLink, ZA; group 2: Acculink carotid, JostentSelfX XF, Luminexx, Omnilink, SMART, sinus-SuperFlex; group 3: Symphony, Wallstent. The signal loss within the stents and the degree of artificial lumen narrowing differed considerably between the stents. Imaging characteristics of group 1 and 2 stents (for classification see Table 1Go) enabled determination of stent patency and accurate delineation of in-stent stenoses by the observers. The MIP algorithm can cause loss of contrast which might artificially worsen the severity of a stenosis or mimic an occlusion. This phenomenon is best appreciated for the group 3 stents.

 
The quality of CEMRA images regarding the overall diagnostic value was ranked to be good (score 1) for all stents and observers.

The observers' performance for visibility of stent patency and in-stent stenoses is summarized in Table 2Go. The scores for the Acculink, DynaLink, JostentSelfX XF, Luminexx, Omnilink, sinus-SuperFlex, SMART, and ZA were equally excellent on both CEMRA source images and MIP images. All observers agreed that visibility of stent patency and in-stent stenoses was worse for the Symphony and Wallstent. Moreover, visibility of patency and in-stent stenoses in these two stents was ranked lower on MIP images compared with the respective source images for 21/24 (88%) of corresponding score pairs. Kappa was 0.51–1.00 (p = 0.027 to p <0.0005) and Spearman's correlation coefficient 0.65–1,00 (p = 0.044 to p <0.0005), indicating high interobserver agreement.


View this table:
[in this window]
[in a new window]
  		Table 2. Subjective and objective evaluation of contrast enhanced MR angiography(CEMRA) images

 
The observers' performance regarding accuracy of grading of in-stent stenosis is given in detail in Table 2Go. For eight (Acculink, DynaLink, JostentSelfX XF, Luminexx, Omnilink, sinus-SuperFlex, SMART, and ZA) out of 10 evaluated stents, the observer's grading differed by less than 10% from the reference standard. For five (Acculink, JostentSelfX XF, Luminexx, SMART, and ZA) of these eight stents, the degree of stenosis was overestimated; for three stents (DynaLink, Omnilink, sinus-SuperFlex) it was underestimated. For the Symphony and Wallstent, the discrepancy between the observers' grading and the reference standard was 22.6% and 33.3%, respectively. For both stents, in-stent stenoses were overestimated. Pearson's correlation coefficients ranged between 0.62 and 0.85 (p = 0.002 to p = 0.059) indicating high interobserver agreement.

Objective data analysis
The magnitude of the in-stent signal intensities and degree of artificial lumen narrowing and, based on these criteria, the group classification are summarized in Table 2Go. The measurements differed considerably between the 10 stents. Eight of the 10 stents (Acculink, DynaLink, JostentSelfX XF, Luminexx, Omnilink, sinus-SuperFlex, SMART, and ZA) exerted less severe artefacts and were consecutively ranked to be partially or well suited for CEMRA. Two stents (Symphony and Wallstent) caused marked image distortion and thus were found to be not well suited for CEMRA. Interestingly, the measurements for the subgroup of eight stents made from nitinol (Acculink, DynaLink, JostentSelfX XF, Luminexx, sinus-SuperFlex, SMART, Symphony, and ZA) were inhomogeneous with the in-stent signal intensity ranging between 34.4% (Symphony) and 95.8% (ZA) and the apparent stent lumen diameter between 43.9% (Symphony) and 77.7% (ZA) relative to the reference tube. The magnitude of artefacts exerted by the stent made from stainless steel (Omnilink: signal intensity within the stent 47.0%, apparent stent lumen diameter 73.6%) was found to be less extensive than that seen in some nitinol stents.


   Discussion
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
References
 
MRA is an alternative to digital subtraction angiography providing several advantages. It is a non-invasive technique without the need for ionizing radiation and potentially harmful iodinated contrast material [6, 7]. In recent years, there have been rapid developments regarding the optimization of MRA sequences. With the introduction of CEMRA, the limitations of time-of-flight and phase contrast techniques in terms of long imaging times and artefacts caused by slow and turbulent flow have widely been overcome [2, 4]. CEMRA is meanwhile well accepted for assessing arterial occlusive disease. However, the monitoring of vascular endoprostheses is still a challenge because of image distortion due to artefacts caused by the stents. According to data in the literature, stent patency can be assessed by CEMRA for a subgroup of stents [20, 21, 23, 24, 27, 28]. A clinically valuable follow-up technique, however, should also enable a reliable detection of critical in-stent stenoses which cause a significant compromise to blood flow. In general, this occurs for stenoses generating a diameter reduction of 50% or more [29, 30]. Consecutively, these stenoses are considered to be relevant for surgical or interventional treatment.

In this in vitro study, 10 stents suitable for the implantation in iliac arteries were evaluated with regard to their imaging characteristics on CEMRA images. Emphasis was placed on subjective criteria (visibility of stent patency and artificial in-stent stenoses, as well as grading of lumen narrowing) and on objective parameters (signal loss within the stent and artificial lumen narrowing). The stents were made from nitinol, stainless steel and a cobalt-based alloy. The quality of images regarding the overall diagnostic value was ranked to be good for all studies. Thus, there was no bias due to a potentially differing image quality.

There are three main types of artefacts associated with metallic vascular implants causing image distortion: (1) radiofrequency (RF) artefacts originating from RF-induced eddy currents in the stent; (2) susceptibility artefacts caused by differences between the magnetic susceptibility of the stent material and the surrounding tissue, leading to local field inhomogeneities; and (3) flow-related artefacts caused by turbulences and consecutive signal loss due to dephasing [9, 12, 14, 16, 27]. The absence of flow and pulsatility in the presented study is a limitation and may reduce the comparability to an in vivo study. However, CEMRA depends on the T1 shortening effects of gadopentetate dimeglumine and is relatively insensitive to flow-related artefacts because of extremely short echo times and thin slices [31]. Thus, the presented setup can be assumed to obtain valid results regarding interstent visibility.

In order to quantify RF induced and susceptibility artefacts, the degree of signal loss inside the stent and artificial lumen narrowing was assessed objectively. According to these measurements, the stents were classified into three groups representing their suitability for MRI. This classification correlated well with the observers' performance regarding the subjective grading of in-stent stenoses. Group 1 and 2 stents (Acculink, DynaLink, JostentSelfX XF, Luminexx, Omnilink, sinus-SuperFlex, SMART, and ZA) were considered to be at least partially suited for CEMRA imaging. For these stents, the observers' accuracy of grading of in-stent stenoses did meet the needs of clinical routine (difference to the standard of reference <10%). For group 3 stents (Symphony and Wallstent), the number of artefacts was more pronounced. Consecutively, the grading of in-stent stenoses was markedly worse, resulting in an overestimation of 23–33%. Hence, implantation of these two stents seems not be favourable if follow-up with CEMRA is intended.

Published data regarding the evaluation of type and amount of artefacts of various stent materials revealed nitinol to be among the favourable alloys and stainless steel to be less suited regarding MR compatibility [12, 13, 15, 16, 32–34]. In the present study, substantial differences between the imaging characteristics within the subgroup of nitinol stents (Acculink, DynaLink, JostentSelfX XF, Luminexx, sinus-SuperFlex, SMART, Symphony and ZA) were noted. This fact emphasises that the magnitude of artefacts not only depends on material composition, but also on other features like wire thickness, total weight and geometry of mesh or slotted tubes. The observation that the Omnilink stent, made from stainless steel, presented in this in vitro study with less severe artefacts than some nitinol stents confirmed that an advantageous design may be able to compensate for disadvantages due to an unfavourable stent material.

For the Symphony and Wallstent, the visibility of stent patency and in-stent stenoses was ranked worse on MIP images than on source images. This was most likely caused by an artificial loss of contrast on MIP images. Loss of contrast is a result of the characteristics of the MIP algorithm which projects noise or distant high signal into the image [26]. MIP images can mimic a stenosis or even a stent or vessel occlusion when the original signal is low, as seen within the Symphony and Wallstent. Therefore, it should be emphasised that for a thorough evaluation of CEMRA studies, interpretation of source and MIP images is mandatory.

In contradiction to the remainder of stents evaluated in this study, the Wallstent presented with a complex morphology of the in-stent signal in terms of bandlike artefacts at both stent endings, a circumscribed region of high signal intensity adjacent to the bandlike artefacts and a severe signal reduction in the central part of the stent. This morphology is the same observed in experimental studies performed by other authors who did not apply in-stent stenoses but evaluated the imaging characteristics of the stent itself [13, 14, 32]. The length of the central region of severe signal reduction observed in the present study exceeded the length of the artificial in-stent stenosis. The signal intensity plots for determination of the apparent in-stent lumen and the region of interest for assessment of in-stent signal intensity were placed within the area of severe signal reduction, but outside the in-stent stenosis.

To our knowledge, there are two studies in literature which evaluated in-stent stenoses on CEMRA images with, however, different setups: on the basis of objective measurements, Letourneau-Guillon et al determined the degree of artificial lumen narrowing and the delineation of a 50% in-stent stenosis in a Luminexx stent (diameter 8 mm), revealing comparable results as observed in our study for this stent [25]. An observer based subjective evaluation was not performed in this study. Maintz et al employed a vascular phantom with an inner tube diameter of 8 mm and evaluated, among others, the Symphony, SMART, Wallstent and ZA stent [35]. The stent-containing tubes were oriented along the main magnetic field B0. The signal loss inside the stent was assessed objectively, the degree of artificial lumen narrowing and the visibility of in-stent stenoses were determined subjectively in a consensus decision. Grading of in-stent stenoses was not performed. As far as comparisons are appropriate, the results are consistent with the data evaluated in the present study. However, the authors observed differing imaging characteristics for the Wallstent in terms of no signal loss inside the stent and a good visibility of the in-stent stenosis. This discrepancy is most likely due to the different orientation of the stent relative to the main magnetic field because it is well documented that the artefact size increases with a larger angle to B0 [9, 12, 13, 15, 25, 36]. This hypothesis is confirmed by another study performed by the same group of authors evaluating the Wallstent at varying angles relative to B0 [14]. According to this latter analysis, the Wallstent caused severe artefacts when the longitudinal axis of the stent was diagonal to B0. Furthermore, two other groups observed similar results as seen in our study in terms of a subtotal to total signal void within the Wallstent even though the stent's orientation was parallel to the main magnetic field [13, 32].

There are limitations to our study. For the sake of comparability, all evaluated stents had a dedicated diameter of 10 mm. Thus, the observed results apply for this specific dimension only. It is possible that the relative artefact size changes with different stent diameters, which might be expected for smaller diameters in particular. However, considering the already outlined data published by Letourneau-Guillon et al and Maintz et al, the imaging characteristics of the Luminexx, Symphony, SMART, and ZA stent were similar for the 10 mm and 8 mm diameter models, indicating low impact of the stent diameter on artefact size [25, 35].

Although the vascular phantom and the experimental setting were designed to simulate the conditions of the iliac vasculature, clinical studies are warranted to investigate if the observations are confirmed in an in vivo setting. First clinical experiences in 27 patients regarding the detection and grading of in-stent stenoses in the JostentSelfX (Abbott Laboratories, Chicago, IL) by CEMRA confirmed the in vitro results of the present study because this stent, which is very similar to the JostentSelfX XF, revealed to be suitable for MR [37]. However, in single cases susceptibility artefacts at the stent's ends simulated significant in-stent stenoses. This phenomenon, which limited the reliability of CEMRA significantly, was not observed under in vitro conditions. Hence, in vitro studies seem to be suitable to roughly classify stents with regard to their MR suitability. However, the final decision has to be made in vivo.

In conclusion, the amount of stent related artefacts differed considerably for the 10 evaluated stents. Two out of 10 tested stents were not suitable for CEMRA follow-up because stent-related artefacts were severe and in-stent stenoses could not reliably be determined. In eight out of 10 stents, artefacts were less pronounced and the observers' performance regarding the determination of stent patency and grading of in-stent stenoses on CEMRA images did meet the requirements of clinical routine. Radiologists should be familiar with this subgroup of stents if follow-up with CEMRA is intended. However, confirmation of the evaluated data under in vivo conditions is mandatory.

Received for publication August 23, 2005. Revision received January 9, 2006. Accepted for publication January 30, 2006.


   References
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 

  1. Borisch I, Horn M, Butz B, Zorger N, Draganski B, Hoelscher T, et al. Preoperative evaluation of carotid artery stenosis: comparison of contrast-enhanced MR angiography and duplex sonography with digital subtraction angiography. AJNR Am J Neuroradiol 2003;24:1117–22.[Abstract/Free Full Text]
  2. Koelemay MJ, Lijmer JG, Stoker J, Legemate DA, Bossuyt PM. Magnetic resonance angiography for the evaluation of lower extremity arterial disease: a meta-analysis. JAMA 2001;285:1338–45.[Abstract/Free Full Text]
  3. Lenhart M, Framme N, Volk M, Strotzer M, Manke C, Nitz WR, et al. Time-resolved contrast-enhanced magnetic resonance angiography of the carotid arteries: diagnostic accuracy and inter-observer variability compared with selective catheter angiography. Invest Radiol 2002;37:535–41.[CrossRef][Medline]
  4. Prince MR, Yucel EK, Kaufman JA, Harrison DC, Geller SC. Dynamic gadolinium-enhanced three-dimensional abdominal MR arteriography. J Magn Reson Imaging 1993;3:877–81.[Medline]
  5. Sueyoshi E, Sakamoto I, Matsuoka Y, Ogawa Y, Hayashi H, Hashmi R, et al. Aortoiliac and lower extremity arteries: comparison of three-dimensional dynamic contrast-enhanced subtraction MR angiography and conventional angiography. Radiology 1999;210:683–8.[Abstract/Free Full Text]
  6. Cochran ST, Bomyea K, Sayre JW. Trends in adverse events after IV administration of contrast media. AJR Am J Roentgenol 2001;176:1385–8.[Abstract/Free Full Text]
  7. Laroche D, Namour F, Lefrancois C, Aimone-Gastin I, Romano A, Sainte-Laudy J, et al. Anaphylactoid and anaphylactic reactions to iodinated contrast material. Allergy 1999;54 Suppl. 58:13–6.
  8. Strotzer M, Lenhart M, Butz B, Volk M, Manke C, Feuerbach S. Appearance of vascular stents in computed tomographic angiography: in vitro examination of 14 different stent types. Invest Radiol 2001;36:652–8.[CrossRef][Medline]
  9. Bartels LW, Smits HF, Bakker CJ, Viergever MA. MR imaging of vascular stents: effects of susceptibility, flow, and radiofrequency eddy currents. J Vasc Interv Radiol 2001;12:365–71.[Medline]
  10. Bartels LW, Bakker CJ, Viergever MA. Improved lumen visualization in metallic vascular implants by reducing RF artifacts. Magn Reson Med 2002;47:171–80.[CrossRef][Medline]
  11. Graf H, Klemm T, Lauer UA, Duda S, Claussen CD, Schick F. [Systematics of imaging artifacts in MRT caused by metallic vascular implants (stents)]. Rofo Fortschr Geb Rontgenstr Neuen Bildgeb Verfahr 2003;175:1711–9.[Medline]
  12. Klemm T, Duda S, Machann J, Seekamp-Rahn K, Schnieder L, Claussen CD, et al. MR imaging in the presence of vascular stents: A systematic assessment of artifacts for various stent orientations, sequence types, and field strengths. J Magn Reson Imaging 2000;12:606–15.[CrossRef][Medline]
  13. Lenhart M, Volk M, Manke C, Nitz WR, Strotzer M, Feuerbach S, et al. Stent appearance at contrast-enhanced MR angiography: in vitro examination with 14 stents. Radiology 2000;217:173–8.[Abstract/Free Full Text]
  14. Maintz D, Kugel H, Schellhammer F, Landwehr P. In vitro evaluation of intravascular stent artifacts in three-dimensional MR angiography. Invest Radiol 2001;36:218–24.[CrossRef][Medline]
  15. Meyer JM, Buecker A, Schuermann K, Ruebben A, Guenther RW. MR evaluation of stent patency: in vitro test of 22 metallic stents and the possibility of determining their patency by MR angiography. Invest Radiol 2000;35:739–46.[CrossRef][Medline]
  16. Wang Y, Truong TN, Yen C, Bilecen D, Watts R, Trost DW, et al. Quantitative evaluation of susceptibility and shielding effects of nitinol, platinum, cobalt-alloy, and stainless steel stents. Magn Reson Med 2003;49:972–6.[CrossRef][Medline]
  17. Hagspiel KD, Leung DA, Nandalur KR, Angle JF, Dulai HS, Spinosa DJ, et al. Contrast-enhanced MR angiography at 1.5 T after implantation of platinum stents: in vitro and in vivo comparison with conventional stent designs. AJR Am J Roentgenol 2005;184:288–94.[Abstract/Free Full Text]
  18. Amano Y, Gemma K, Kawamata H, Kumazaki T. Fat-suppressed gadolinium-enhanced three-dimensional magnetic resonance angiography adequately depicts the status of iliac arteries following atherectomy and stent placement. Cardiovasc Intervent Radiol 1998;21:345–7.[CrossRef][Medline]
  19. Cavagna E, Berletti R, Schiavon F. In vivo evaluation of intravascular stents at three-dimensional MR angiography. Eur Radiol 2001;11:2531–5.[CrossRef][Medline]
  20. Juergens KU, Tombach B, Reimer P, Vestring T, Heindel W. Three-dimensional contrast-enhanced MR angiography of endovascular covered stents in patients with peripheral arterial occlusive disease. AJR Am J Roentgenol 2001;176:1299–303.[Abstract/Free Full Text]
  21. Link J, Steffens JC, Brossmann J, Graessner J, Hackethal S, Heller M. Iliofemoral arterial occlusive disease: contrast-enhanced MR angiography for preinterventional evaluation and follow-up after stent placement. Radiology 1999;212:371–7.[Abstract/Free Full Text]
  22. Matsumoto AH, Teitelbaum GP, Barth KH, Carvlin MJ, Savin MA, Strecker EP. Tantalum vascular stents: in vivo evaluation with MR imaging. Radiology 1989;170:753–5.[Abstract/Free Full Text]
  23. Schurmann K, Vorwerk D, Bucker A, Neuerburg J, Grosskortenhaus S, Haage P, et al. Magnetic resonance angiography of nonferromagnetic iliac artery stents and stent-grafts: a comparative study in sheep. Cardiovasc Intervent Radiol 1999;22:394–402.[CrossRef][Medline]
  24. Tello R, Thomson KR, Witte D, Becker GJ, Tress BM. Dynamic gadolinium DTPA-enhanced magnetic resonance of intravascular stents. Invest Radiol 1998;33:411–4.[CrossRef][Medline]
  25. Letourneau-Guillon L, Soulez G, Beaudoin G, Oliva VL, Giroux MF, Qin Z, et al. CT and MR imaging of nitinol stents with radiopaque distal markers. J Vasc Interv Radiol 2004;15:615–24.[Medline]
  26. Anderson CM, Saloner D, Tsuruda JS, Shapeero LG, Lee RE. Artifacts in maximum-intensity-projection display of MR angiograms. AJR Am J Roentgenol 1990;154:623–9.[Free Full Text]
  27. Amano Y, Ishihara M, Hayashi H, Gemma K, Kawamata H, Amano M, et al. Metallic artifacts of coronary and iliac arteries stents in MR angiography and contrast-enhanced CT. Clin Imaging 1999;23:85–9.[CrossRef][Medline]
  28. Matsumoto AH, Teitelbaum GP, Carvlin MJ, Barth KH, Savin MA, Strecker EP. Gadolinium enhanced MR imaging of vascular stents. J Comput Assist Tomogr 1990;14:357–61.[Medline]
  29. May AG, van de Berg L, Deweese JA, Rob CG. Critical arterial stenosis. Surgery 1963;54:250–9.[Medline]
  30. Rose SC. Noninvasive vascular laboratory for evaluation of peripheral arterial occlusive disease: Part I--hemodynamic principles and tools of the trade. J Vasc Interv Radiol 2000;11:1107–14.[Medline]
  31. Prince MR. Gadolinium-enhanced MR aortography. Radiology 1994;191:155–64.[Abstract/Free Full Text]
  32. Hilfiker PR, Quick HH, Debatin JF. Plain and covered stent-grafts: in vitro evaluation of characteristics at three-dimensional MR angiography. Radiology 1999;211:693–7.[Abstract/Free Full Text]
  33. Barras CD, Myers KA. Nitinol - its use in vascular surgery and other applications. Eur J Vasc Endovasc Surg 2000;19:564–9.[CrossRef][Medline]
  34. Duerig T, Tolomeo D, Wholey M. An overview of superelastic stent design. Min Invas Ther Allied Technol 2000;9:235–46.
  35. Maintz D, Tombach B, Juergens KU, Weigel S, Heindel W, Fischbach R. Revealing in-stent stenoses of the iliac arteries: comparison of multidetector CT with MR angiography and digital radiographic angiography in a Phantom model. AJR Am J Roentgenol 2002;179:1319–22.[Abstract/Free Full Text]
  36. Trost DW, Zhang HL, Prince MR, Winchester PA, Wang Y, Watts R, et al. Three-dimensional MR angiography in imaging platinum alloy stents. J Magn Reson Imaging 2004;20:975–80.[CrossRef][Medline]
  37. Hamer OW, Finkenzeller T, Borisch I, Paetzel C, Zorger N, et al. In vivo evaluation of patency and in-stent stenoses after implantation of nitinol stents in iliac arteries using MR angiography. AJR Am J Roentgenol 2005;185:1282–8.[Abstract/Free Full Text]




This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hamer, O W
Right arrow Articles by Zorger, N
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hamer, O W
Right arrow Articles by Zorger, N

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
BJR DMFR IMAGING  ALL BIR JOURNALS