| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Experimental Studies |
1 From the Department of Radiology-MRI, Bolwell B 124, University Hospitals of Cleveland, Case Western Reserve University, 11100 Euclid Ave, Cleveland, OH 44106 (F.K.W., M.W., J.S.L.); Department of Radiology, Benjamin Franklin University Hospital, Free University, Berlin, Germany (F.K.W., K.R., K.J.W.); Schering, Berlin, Germany (W.E.); and Siemens, Erlangen, Germany (M.W.). From the 2000 RSNA scientific assembly. Received November 12, 2001; revision requested January 28, 2002; final revision received May 1; accepted June 5. Supported by grants R33 CA88144-01 and R01 CA81431-02 from the National Cancer Institute and grants from Siemens Medical Solutions. Address correspondence to F.K.W. (e-mail: wackerfrank@web.de).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: All experiments were performed with MR imaging at 0.2 T. MR image–guided interventions were performed in USPIO-enhanced vessels in four pigs. With near real-time MR image guidance (acquisition time, 0.64 second per section), the splenic and renal arteries were consecutively catheterized by using a susceptibility artifact–based catheter–guide wire combination. Angioplasty and stent implantation were performed four times in the renal artery and twice in the iliac artery. Intraaortal signal intensity (SI) was measured during the interventions.
RESULTS: After administration of SH U 555 C (40 µmol of iron per kilogram of body weight), a three-dimensional MR angiographic sequence was performed that allowed visualization of the abdominal and pelvic vessels that were as small as 2 mm in diameter. Catheterization, angioplasty, and stent implantation were successfully guided in the USPIO-enhanced vasculature. Sixty minutes after contrast agent injection, the mean aortic SI was 70% of the maximum measured enhancement levels.
CONCLUSION: One intravenous injection of SH U 555 C enabled long, continuous intravascular SI enhancement at MR angiography, and, in combination with susceptibility artifact–based device tracking, the injection allowed the performance of MR imaging–guided intravascular interventions in an open MR imaging system.
© RSNA, 2003
Index terms: Animals • Interventional procedures, experimental studies, 954.1268, 961.1268, 988.1268 • Magnetic resonance (MR), contrast media, 954.129412, 954.12943, 961.129412, 961.12943, 988.129412, 988.12943 • Magnetic resonance (MR), vascular studies, 954.129412, 954.12943, 961.129412, 961.12943, 988.129412, 988.12943 • Stents and prostheses, 961.1286
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Compared with fluoroscopic guidance, magnetic resonance (MR) image guidance can provide superb soft-tissue contrast to visualize the vessel lumen, the vessel wall, and the surrounding structures, and this guidance can provide flow velocity data without any radiation exposure (1–3). MR imaging contrast agents may be used to enhance the visualization of blood vessels. During vascular interventional procedures, the commonly used extracellular gadolinium-based preparations require repeated injections because of their very short intravascular retention time. The maximal approved dose of these preparations can be easily achieved during an interventional procedure (4), and the associated enhancement of the background tissue can reduce the vessel contrast and thus the overall image quality.
For prolonged contrast enhancement of blood vessels, the use of intravascular blood pool agents, such as macromolecular gadolinium-based preparations and ultrasmall superparamagnetic iron oxide (USPIO), has been proposed (5–10). Since open MR imaging systems usually operate at low field strength, USPIOs are particularly suited as intravascular contrast agents, because compared with gadolinium chelates, their relaxivity r1 increases exponentially with decreasing Larmor frequency of protons (11).
The goal of this study was to evaluate the feasibility of using a USPIO (SH U 555 C) as an intravascular contrast agent for MR image–guided vascular procedures in an open MR imaging system.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
A three-dimensional (3D) MR angiographic sequence was used along with section interpolation, which allowed thin-section MR angiography in the low-field-strength system (spoiled gradient-echo [GRE] sequence, repetition time msec/echo time msec, 6.3/2.4; flip angle, 25°). This sequence was performed in a high-spatial-resolution mode, with a section thickness of 1.3 mm, four acquisitions, in-plane resolution of 1.4 x 0.8 mm, and acquisition time of 3 minutes 21 seconds, and in a breath-hold mode, with section thickness of 2 mm, one acquisition, in-plane resolution of 1.9 x 0.8 mm, and acquisition time of 20 seconds.
For interventional procedures, guidance was provided with a rapid two-dimensional (2D) spoiled GRE sequence with a minimum echo time of 2 msec and a minimum repetition time of 5 msec. The total acquisition time was 0.64 second per section (128 x 256 matrix, 6–8-mm section thickness, 25° flip angle). The maximum intensity projections with both MR angiographic modes were compared on the basis of the visualization of the renal arteries by two readers (F.K.W., K.R.) in consensus to identify the better overall image quality. In addition, the diameter of the smallest clearly visible artery was measured for both MR angiographic modes and the rapid 2D GRE sequence by one author (F.K.W.). With the standard software provided on the MR imager, measurements were determined from inner vessel wall to inner vessel wall with electronic calipers on images that were magnified three times.
For situations that required an anatomic overview during the interventional procedure, we performed breath-hold 3D MR angiography for vascular imaging and 2D spoiled GRE MR imaging (110/9; flip angle, 80°; sections, five; section thickness, 8 mm; measurement time, 23 seconds) for section visualization.
MR Angiography–guided Interventions
The interventions were performed by one of us (F.K.W.) in four farm pigs (body weight, 32–42 kg). The pigs were anesthetized with 2 mL of ketamine 10% (Mallinckrodt Vet, Burgwedel, Germany) administered intravenously every 20–30 minutes and 2 mg of midazolam (Dormicum; Roche, Grenzach, Germany) administered every 60 minutes. The experimental protocol was approved by the appropriate governmental regulatory committee.
After the animal was positioned in the MR imager, 3D MR angiograms were obtained before and after a bolus injection of SH U 555 C (Schering, Berlin, Germany). This USPIO consists of carboxydextran-coated iron oxide particles with a mean size of 20 nm. A dose of 40 µmol of iron per kilogram of body weight, which is the highest approved dose for SH U 555 C, was used in all animals. The maximum intensity projections of the MR angiograms allowed an overview of the abdominal vasculature for planning of the MR imaging–guided procedures. The 2D GRE images used for procedure guidance were continuously displayed in the magnet room and were used to enable visualization of the aorta and the iliac, renal, and splenic arteries. With this sequence, the devices were passed through the introducer sheath located in the carotid artery (n = 1) or the distal iliac artery (n = 3) and advanced into the target vessels with MR imaging guidance.
Intravascular Devices and Procedures
All intravascular devices were visualized by means of their susceptibility artifacts. For catheter manipulations, we used a fully MR-visible prototype 5-F-diameter catheter (Somatex, Berlin, Germany) constructed with ferrite admixture that was designed for susceptibility artifact–based visualization. With the 2D GRE sequence we used for MR imaging guidance, the MR imaging catheter artifact was 8.4 mm ± 1.3 in diameter at a 90° angle relative to the constant magnetic induction field (B0). We observed smaller artifact diameters in situations in which the angle relative to B0 was less than 90°, which was in conformity with previous experimental data (12).
For over-the-wire manipulations, we used a steerable hydrophilic-coated 35-inch guide wire (Radifocus; Terumo, Tokyo, Japan), which generates no MR image artifact, and a prototype guide wire (Ferro Tip; Somatex). The prototype guide wire consisted of a 35-inch tapered MR imaging artifact–free nitinol shaft and a 2-mm ferromagnetic tip that induces a 14-mm-diameter round signal void with the 2D GRE sequence. Both the mechanical and biologic properties of the catheter and prototype guide wire used in this study do not differ from those of conventional angiographic devices.
To test the feasibility of MR angiography–guided vascular interventions, an interventional task was delineated that consisted of catheterization of the splenic and the right renal arteries, angioplasty and stent implantation in the renal artery, and stent implantation in the iliac artery. The catheter passes were performed to demonstrate the ability of MR image guidance to help detect deviations of the catheter and to help correct them by means of rotation of the catheter or insertion of a guide wire. In all animals, following successful catheterization of the renal artery orifice, a guide wire was passed through the renal artery, and the catheter was exchanged for an angioplasty catheter over the guide wire. The balloon was inflated with the diluted gadolinium-based contrast agent, gadopentetate dimeglumine (Magnevist; Schering), in a concentration of 0.05 mol/L.
A stent was then positioned in the ostial portion of the renal artery, with its proximal end at the level of the aortic wall. Two animals additionally underwent angioplasty and stent implantation in the iliac artery 5 mm distal to the aortal bifurcation. For angioplasty, balloon catheters (Ultra Thin; Boston Scientific, Watertown, Mass) with 4- and 8-mm diameters were available and were supplied with additional MR-visible anterior and posterior markers. For stent implantation, we used balloon-expandable stents (Palmaz; Cordis, Miami, Fla) that were 4–9 mm in diameter and 11.6–19.6 mm in length in two animals and self-expandable stents (Symphony; Boston Scientific) that were 5–8 mm in diameter and 20 mm in length in four animals. After finishing the MR imaging–guided interventions, the stent positions were checked angiographically by using digital subtraction angiography with a unit (Integris V3000; Phillips, Best, the Netherlands).
During all interventional procedures, the time required to perform the procedure was recorded, along with the time point when the intravascular signal intensity (SI) was no longer adequate according to the examiner’s (F.K.W.) subjective assessment. For numeric analysis, aortic SI enhancement (E) was calculated with the following equation: E = (SIpost - SIpre) x 100 /SIpre, where SIpost indicates SI after contrast material injection and SIpre indicates SI before contrast material injection. During the interventions, the enhancement was measured retrospectively by using fast 2D GRE MR images. A circular region of interest was placed within the center of the aortic lumen 1 cm above the renal artery, and SIs of regions of interest were measured by one of us (K.R.) at the corresponding location on the images obtained during the interventions. The size of the region of interest varied according to the target artery and ranged from 5.6 to 8.7 mm (mean, 7.3 mm) in diameter. Since 3D MR angiography was the first sequence performed after contrast agent injection, no enhancement values could be obtained immediately after injection.
For each procedure, the diameter of the target vessel was measured on the 3D and 2D MR angiograms as described previously, and the stent position relative to the aortic wall at the renal artery ostium and the aortic bifurcation was measured by one of us (F.K.W.) on the conventional angiogram by using the standard software provided on the angiographic unit. A one-sample t test was used to evaluate whether the stent deviation determined with MR image guidance was no larger than the 2-mm precision that can usually be achieved by using fluoroscopy (13). A one-tailed P value of less than .05 was considered to indicate a statistically significant difference.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
|
|
|
|
|
|
|
|
|
|
The escape of contrast material–enhanced blood into the retroperitoneum was immediately identified at near real-time imaging (Fig 3). Because a covered stent was not available, the hemorrhaging was stopped by using balloon occlusion with MR image guidance, and a balloon-expandable stent was implanted to overlay the ruptured site (Fig 3c). Hematoma growth stopped immediately after stent implantation. The animal survived this complication for 90 minutes and was then sacrificed at the end of the experiment. Stent implantation succeeded in all other experiments as well. Fluoroscopic monitoring of the stent position revealed deviations of 2–13 mm (mean, 7.3 mm) from the expected stent position. This deviation was significantly greater (P = .015) than the 2-mm deviation currently accepted with fluoroscopic guidance.
All catheter manipulations and interventions were performed in 90 minutes or less (range, 54–90 minutes). In two animals, the intravascular SI was subjectively considered inadequate after 73 and 85 minutes. The time courses of the SI during the experiments demonstrated an exponential decline in intravascular SI during the interventional procedures (Fig 4). At 60 minutes after the start of contrast agent injection, the mean SI was 70% (327.8/465.3) of the maximum measured levels.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
In general, contrast-enhanced MR angiography is performed by using gadolinium chelates as the contrast agents. USPIOs are an alternative because their r1 is four to five times higher than that of gadopentetate dimeglumine (measured at 0.47 T) (18) and is even greater at lower field strengths (11). Moreover, USPIOs enable substantially longer SI enhancement of vessels than do gadolinium polymers (10,19,20). These properties make USPIOs theoretically suitable for intravascular contrast enhancement during MR image–guided interventions with low-field-strength systems, which was verified by findings in our study. The elevation in SI after contrast agent injection and the duration of intravascular enhancement were sufficient for the acquisition of a high-spatial-resolution MR angiogram with visualization of vessels as small as 2 mm in diameter and a good anatomic overview.
On the basis of these ideas, targeted vessel selection and straightforward interventional procedures could be performed with near real-time MR image guidance in straight vessels larger than 4 mm in diameter. The high intravascular SI permitted excellent visualization and MR image–guided positioning of the susceptibility artifact–based catheters and guide wires. Changes in catheter position and dislodgment from the selected vessel were reliably detected and prevented in our experiments. In a study by Manke and colleagues (21), stents were implanted in nonopacified vessels by using a landmark technique. They described stent misplacement caused by patient movement and subintimal recanalization. These complications might have been prevented with use of intravascular contrast agents, specially designed devices, and near real-time imaging used in our study.
The susceptibility artifact–based catheter used in our study allowed visualization of the whole device at MR imaging, which was similar to the visualization possible with conventional x-ray techniques. However, the size of the artifact was greater than the actual catheter width, which is a problem in smaller vessels, where the signal void of the device may overlay the vessel wall. Different catheters with smaller artifact induction must be developed to allow reliable catheterization of smaller vessels. The same is true for the guide wire tip, which has been designed to avoid dislocation during over-the-wire procedures, such as angioplasty and stent implantation. Options that also need to be considered to change artifact size include the use of shorter echo times of the real-time sequence, which was not possible with the current gradient strength of our MR imager, and the use of other pulse sequences (12).
The major drawback of the current study is the reduced spatial and temporal resolution in a low-field-strength system. Along with the inaccuracy associated with susceptibility artifact–based devices, the reduced spatial and temporal resolution was responsible for most of the stent deviation in this study. This inaccuracy would not be acceptable for clinical renal artery stent implantation. Interventional procedures that require such a high precision or that are performed in smaller or curved vessels require a further improvement in both spatial and temporal resolution. This seems to be easier to achieve with high-field-strength MR imagers with powerful gradient systems than with the low-field-strength system used in this study (21–23).
However, patient access is still difficult with conventional closed-bore magnets, and many interventional devices present considerable safety problems when used for imaging with a high-field-strength system (24–26). In view of the results presented in our study and the drawbacks regarding the use of current high-field-strength MR imaging systems, improvement of the currently available open low-field-strength imagers with a C-arm configuration is a real alternative even for vascular interventional procedures, not least because of the specific advantages of the use of USPIOs at a lower field strength as demonstrated in this investigation.
Practical application: The strength and persistence of intravascular SI enhancement after SH U 555 C injection indicate that it is a potentially valuable contrast agent for MR image–guided vascular procedures.
|
FOOTNOTES |
|---|
Author contributions: Guarantors of integrity of entire study, F.K.W., K.J.W.; study concepts, F.K.W., J.S.L.; study design, K.R., W.E.; literature research, K.R., W.E.; experimental studies, F.K.W., K.R., W.E.; data acquisition, F.K.W., K.R.; data analysis/interpretation, F.K.W., M.W.; manuscript preparation, F.K.W., J.S.L.; manuscript definition of intellectual content, F.K.W., K.J.W.; manuscript editing, F.K.W., M.W.; manuscript revision/review, K.J.W., J.S.L.; manuscript final version approval, all authors.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. J. Krueger, P. Ewert, S. Yilmaz, D. Gelernter, B. Peters, K. Pietzner, A. Bornstedt, B. Schnackenburg, H. Abdul-Khaliq, E. Fleck, et al. Magnetic Resonance Imaging-Guided Balloon Angioplasty of Coarctation of the Aorta: A Pilot Study Circulation, February 28, 2006; 113(8): 1093 - 1100. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. A. Omary, J. A. Gehl, B. E. Schirf, J. D. Green, B. Lu, F. S. Pereles, J. Huang, A. C. Larson, and D. Li MR Imaging- versus Conventional X-ray Fluoroscopy-guided Renal Angioplasty in Swine: Prospective Randomized Comparison Radiology, February 1, 2006; 238(2): 489 - 496. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. K. Wacker, D. Elgort, C. M. Hillenbrand, J. L. Duerk, and J. S. Lewin The Catheter-Driven MRI Scanner: A New Approach to Intravascular Catheter Tracking and Imaging-Parameter Adjustment for Interventional MRI Am. J. Roentgenol., August 1, 2004; 183(2): 391 - 395. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
