Neuroradiology (2011) 53:449–456 DOI 10.1007/s00234-010-0744-z
INTERVENTIONAL NEURORADIOLOGY
MR imaging of hydrogel filament embolic devices loaded with superparamagnetic iron oxide or gadolinium Monika Killer & Edward M. Keeley & Gregory M. Cruise & Anne Schmitt & Mark R. McCoy
Received: 27 January 2010 / Accepted: 28 June 2010 / Published online: 13 July 2010 # Springer-Verlag 2010
Abstract Introduction We evaluated hydrogel filaments loaded with barium sulphate and either gadolinium or superparamagnetic iron oxide (SPIO) in an effort to develop an embolic material that is visible with fluoroscopic and magnetic resonance imaging. Methods Hydrogel filaments were prepared with gadolinium and iron concentrations ranging from 1,500 to 7,500 and 500 to 2,500 ppm, respectively. The filaments were encased in agar and imaged using an MR scanner. Embolisation of eight aneurysms (seven bifurcation, one sidewall) in seven rabbits was performed using hydrogel filaments loaded with gadolinium (n=4) or SPIO (n=4). Angiographic evaluations occurred immediately post-treatment and at 13 weeks. Magnetic resonance angiography (MRA) evaluations occurred immediately post-treatment or 13 weeks post-treatment. Results Based on the in vitro results, we selected 4,500 and 2,000 ppm for gadolinium and iron loadings, respectively, for the in vivo experiments. Loading the filaments with gadolinium or SPIO did not affect the angiographic results, as M. Killer (*) : A. Schmitt Neuroscience Institute, Christian Doppler Clinic, Paracelsus Medical University, Ignaz Harrer Strasse 79, 5020 Salzburg, Austria e-mail:
[email protected] E. M. Keeley : G. M. Cruise MicroVention Terumo, 1311 Valencia, Tustin, CA 92780, USA M. R. McCoy Departments of Radiology and MRI, Christian Doppler Clinic, Paracelsus Medical University, Ignaz Harrer Strasse 79, 5020 Salzburg, Austria
embolic masses were readily evident with some distinguishing of individual filaments. In MRA, the hydrogel filaments loaded with SPIO were hypointense, and the hydrogel filaments loaded with Gd were hyperintense. The hyperintensity of the Gd-loaded filaments confounded the ability to distinguish between flow and the embolic devices. The hypointensity of the hydrogel filaments loaded with SPIO provided sufficient contrast between the embolic devices and the blood flow to allow of aneurysm occlusion evaluation using MRA. Conclusion Based on these results, we are focusing on loading hydrogel filaments with SPIO in an effort to provide adequate visualisation for use in MR-guided interventions. Keywords Magnetic resonance angiography . Therapeutic embolisation . Aneurysm . Hydrogel
Introduction Embolisation is a widely accepted technique for the treatment of intracranial aneurysms to prevent rupture or rerupture. The vast majority of these interventions are performed using fluoroscopic guidance, while other imaging techniques, such as magnetic resonance angiography (MRA), are used in conjunction with angiography in follow-up evaluations. The use of MRA [1, 2] in follow-up is increasing as a result of the less invasive nature of the examinations, multiplanar capability and soft tissue contrast. Additionally, MRA does not use ionizing radiation and provides physiological as well as anatomical information. The advantages of MRA that are driving its increased use for follow-up would also be beneficial during the intervention. While improvements in spatial and temporal
450
imaging are required, new generations of MR scanners and real-time sequences have made MR-guided interventions possible [3] resulting in preclinical MR-guided interventions in a wide variety of indications [4–7]. However, MR-guided neurointervention is still experimental at this stage, as indicated by the lack of imaging sequences and literature reports. The major limitation for MR-guided interventions appears to be the lack of compatible interventional tools [8]. While substantial work is being performed in the development of catheters and guidewires that are MR compatible [9, 10], less research is being conducted on MR-visible embolic devices. While MR follow-up of embolised intracranial aneurysms is widely performed [2], the artefacts from the platinum coils confound the interpretation of the images in a small number of cases [11, 12]. Additionally, a rim of increased intensity adjacent to a 2- to 3-mm area of signal void and obscuration surrounding the brain parenchyma has been observed after embolisation with platinum coils [13]. Furthermore, platinum coils do not create susceptibility artefacts and thus are only visible through the effect they have on the flow. As a result, single loops of platinum coils are not visible during MRA. While this is acceptable for follow-up, it precludes MR-guided interventions. In an effort to overcome these and other limitations of platinum coils, we are evaluating metal-free, hydrogel filament embolic devices. These hydrogel filaments have helical diameters and lengths comparable to those of platinum coils and are delivered in a similar manner. Previously, we have reported our investigations into the angiographic/histologic durability [14] and artefact-free computed tomography angiography evaluations [15] of experimental aneurysms embolised with hydrogel filaments. However, these hydrogel filaments were not visible when imaged with MR. In an effort to impart MR visibility, gadolinium and superparamagnetic iron oxide (SPIO) particles were incorporated into the hydrogel filaments. In this experiment, we evaluated the MR visibility of these hydrogel filaments in vitro and in vivo.
Materials and methods Preparation of the hydrogel filaments Gadolinium DTPA methacrylate Gadolinium diethylenetriaminepentaacetic acid (Gd-DTPA; 2.7 g, 0.005 mol) was dissolved in 95 ml water. 3-(Dimethyl amino)propyl-3ethylcarbodiimide hydrochloride (2.1 g, 0.011 mol) and aminoethyl methacrylate (1.65 g, 0.01 mol) were added to the Gd-DTPA solution. The pH was adjusted to 8.0 with triethylamine, and the reaction was stirred for 5 h. The
Neuroradiology (2011) 53:449–456
water was removed by rotary evaporation, and the GdDTPA methacrylate was recovered. Hydrogel filaments Detailed descriptions of the preparation, characterization and physical evaluation of the hydrogel filaments are given elsewhere [16]. For the in vitro experiments, the gadolinium containing filaments comprised 65% (w/w) barium sulphate, 18% poly(ethylene glycol) and 11% acrylic acid supplemented with 1.4–7% Gd-DTPA methacrylate (1,500–7,500 ppm Gd). The SPIO-containing filaments comprised 67% (w/w) barium sulphate, 19% poly(ethylene glycol) and 11% acrylic acid supplemented with 0.1–0.6% 10-μm SPIO particles (500 to 2,500 ppm iron). Based on the in vitro imaging results, hydrogel filaments were prepared for the in vivo experiment. The gadolinium containing filaments comprised 65% (w/w) barium sulphate, 18% poly (ethylene glycol) and 11% acrylic acid supplemented with 4% Gd-DTPA methacrylate. The SPIOcontaining filaments comprised 67% (w/w) barium sulphate, 19% poly(ethylene glycol) and 11% acrylic acid supplemented with 0.5% 10-μm SPIO particles. The prepolymer solution of the above components was injected into a Hytrel® (Dupont, Wilmington, DE, USA) extrusion with an inner diameter of 0.018 in. In the in vivo experiment, the polymerization tubing was wrapped around a 5-mm mandrel to impart a 5-mm helical diameter to the hydrogel filament. After polymerization, the Hytrel tubing was dissolved in chloroform, leaving the hydrogel filament. After washing and drying, the hydrogel filament was trimmed to 5, 10 or 15 cm lengths, loaded into an introducer with a LuerLok hub, packaged and sterilised using gamma irradiation. For both the in vitro and in vivo experiments, actual gadolinium and iron concentrations of the hydrogel filaments were determined using inductively coupled plasma–mass spectrometry. In vitro study The gadolinium and iron oxide-loaded filaments were placed in a petri dish. While taking care to prevent the introduction of air bubbles, the petri dish was completely filled with a 2% agar solution that was heated to 80°C. The petri dishes were placed in a refrigerator to solidify the agar. The hydrogel filaments were imaged using a 3-T MR scanner (Achieva, Philips Medical Systems, Best, The Netherlands). Table 1 summarises the MR examination parameters for the in vitro study. In vivo study The rabbits were cared for in accordance with the Austrian regulations governing animal experiments, and the experimentation was approved by the committee for animal experiments from Land Salzburg, Austria. Eight experimental aneurysms (seven bifurcation, one sidewall) were created in seven New Zealand white rabbits. For all
Neuroradiology (2011) 53:449–456
451
Table 1 MR-imaging sequence parameters. Study
Weighting
Type
Slice plane
TR (ms)
TE (ms)
Flip angle
In vitro
T1 T2 – T1 T1 T2 –
SE TSE 3D TOF SE SE TSE 3D TOF
TRA COR TRA COR COR TRA SAG TRA TRA
450–650 3,000–5,000 23 450–650 Shortest 3,000–5,000 23
13 80 3.45 13 4.6 80 3.45
90 90 18 90 80 90 18
In vivo
Matrix
FOV (mm)
Pixel size (mm)
Slice thickness (mm)
132×86 208×130 368×244 132×86 260×199 208×130 368×244
120×97 120×104 200×200 120×97 150×143 120×104 200×200
0.9×1.12 0.57×0.72 0.54×0.82×0.7 0.9×1.12 0.57×0.72 0.57×0.72 0.54×0.82×0.7
2 2 – 2 3 1 –
TRA transverse, COR coronal, SAG sagittal, SE spin-echo, TSE turbo spin-echo, 3D TOF three-dimensional time of flight, FOV field of view
operative procedures, the animals were anaesthetised using intramuscular injection of 20–30 mg/kg ketamine hydrochloride and 0.2 mL 2% xylazine hydrochloride, followed by maintenance anaesthesia by intravenous injection of a saline solution of ketamine and xylazine (5:1:5; 0.5–1 mL/h/kg). The microsurgical construction of the aneurysms was performed according to previously described methods [17]. Pretreatment angiography was performed to document the aneurysm’s width, length and neck. Since the aneurysm was constructed from a cylindrical vein, the aneurysm volume was calculated from the width and length determinations, assuming cylindrical geometry (π×width2 ×length/4). Embolisation of the aneurysms was performed with hydrogel filaments containing gadolinium or hydrogel filaments containing SPIO. For these initial feasibility evaluations, four aneurysms per group were used. Four was selected to provide opportunity to use approximately ten embolic devices per group. Additionally, our previous work with CT imaging demonstrated strong concordance between subjects, supporting lower numbers of replicates per group [15]. Although the ultimate goal of this work is the development of embolic devices suitable for use with MR guidance, we chose to embolise under fluoroscopic guidance (Siemens, Multistar) to permit evaluation of the embolic devices while we develop suitable imaging sequences for preclinical neurointerventions. For this initial experiment, the hydrogel filaments were injected through a 0.019- or 0.021-in. microcatheter into the aneurysm sac using the technique for the Berenstein Liquid Coils (Boston Scientific, Fremont, CA, USA) or Jet coils (Dendron/eV3, Bochum, Germany). The length of the hydrogel filament was selected by the interventionalist (M.K.) in light of the expected amount of space remaining in the aneurysm. The introducer was flushed with saline to remove the air. The tip of the introducer was seated into the hub of the microcatheter, and the hydrogel filament was injected into the microcatheter and subsequently into the aneurysm sac. After performing final angiography to document aneurysm occlusion, all catheters and the sheath were removed. Angiographic evaluations occurred post-
treatment and at the 13-week follow-up. Angiographic occlusion was evaluated post-treatment and at follow-up according to the Raymond scale [18] as complete, near complete or incomplete. At the conclusion of the experiment, the rabbits were sacrificed by a lethal overdose of the anaesthetic agents and 2 mL of T61® euthanasia solution (Veterinaria AG, Zürich, Switzerland). MR imaging Initial unenhanced MRA was performed using a 3-T scanner (Achieva, Philips Medical Systems) with the parameters described in Table 1. Post-processing was performed on a workstation (View Forum, Philips Medical Systems) for viewing multiplanar reformatted images. Three-dimensional maximum intensity projection (MIP) views were prepared. For the in vitro imaging, the agar plates were placed on a bag of saline inside a 16-channel neurovascular array coil (Sense Neurovascular Coil NV16, Philips) wrapped around the stack. For the in vivo experiment, the rabbit was placed inside the 16-channel neurovascular array coil. To be able to detect changes in MR visibility of the hydrogel filaments over time, two aneurysms from each group were imaged immediately posttreatment and the remaining two aneurysms were imaged immediately before sacrifice at 13 weeks post-treatment. Evaluation criteria For the in vitro experiment, the effect of the concentration of gadolinium and SPIO on MR Table 2 Nominal and actual iron and gadolinium contents. Iron content Nominal (ppm) 500 1,000 1,500 2,000 2,500
Gadolinium content Actual (ppm)
Nominal (ppm)
Actual (ppm)
664 1,590 1,980 3,720 3,540
1,500 3,000 4,500 6,000 7,500
1,270 2,310 3,280 4,350 5,850
452
Neuroradiology (2011) 53:449–456
development of the hydrogel filaments or the embolisation of the rabbit aneurysms. The reported scoring represents the consensus of the authors. The angiography images were scored in a blinded manner. Due to the different appearances of the Gd and SPIO-loaded hydrogel filaments, the MRA images could not be scored blindly.
Results Hydrogel filaments The theoretical and experimentally determined loadings of gadolinium or iron in the hydrogel filaments are shown in Table 2. In vitro study Figure 1 shows the appearance of gadolinium and SPIOloaded hydrogel filaments encased in agar using T1-weighted, T2-weighted and 3D TOF imaging. The gadolinium-loaded filaments are hyperintense and the SPIO-loaded filaments are hypointense. Both types of hydrogel filaments are readily apparent in T1-weighted and 3D TOF imaging but are difficult to distinguish using T2-weighted imaging.
Fig. 1 In vitro MR imaging of hydrogel filaments. a Photographic image of an agar plate with hydrogel filaments loaded with gadolinium and SPIO. From left to right, the loading concentrations of the hydrogel filaments are 4,820 ppm Gd, 2,740 ppm Gd, 2,100 ppm Fe and 1,370 ppm Fe. b Frontal, T1-weighted MR image of agar plate. c Frontal, T2-weighted MR image of agar plate. d Frontal, 3D TOF MR image of agar plate
In vivo study
visibility was evaluated subjectively. For the in vivo experiment, the images from angiography and MRA were scored for visibility of the embolic mass, visibility of individual filaments, visibility of residual flow in the aneurysm neck and visibility of residual flow in the aneurysm sac as none, unacceptable, acceptable or excellent. The presence of artefacts on the angiography and MRA images were scored as none, minor or major. Two of the co-authors (M.K. and M.R.M.) scored the angiographic and MRA images. One of the two co-authors (M.R.M.) who performed the scoring was not involved in the
Procedural Table 3 summarises the procedural results from the eight aneurysms in this study. Both the Gd and SPIOloaded hydrogel filaments were loaded into a microcatheter using injection and then flushed with saline through the microcatheter and into the aneurysm sac. Both were easily delivered into the aneurysm sac using the same technique as for other injectable embolics. The filaments readily packed into the aneurysm sac, although parent artery incursions were unavoidable due to the injection delivery.
Table 3 In vivo procedural results. Subject
SPIO
Gd
1 2 3 4 5 6 7 8
Aneurysm width (mm)
Aneurysm length (mm)
Aneurysm neck (mm)
Aneurysm volume (cm3)
Number of devices implanted
Total device length (cm)
4.8 4.6 3.5 3.8 6.5 4.4 3.8 3.8
8.2 8.6 6.8 5.1 8.2 5.9 7.1 4.6
3.5 4.4 2.8 2.0 5.7 2.1 3.6 2.1
0.15 0.14 0.07 0.06 0.27 0.09 0.08 0.05
2 3 2 1 7 2 1 3
20 35 15 10 50 10 10 20
Neuroradiology (2011) 53:449–456
453
Table 4 In vivo imaging results. Angiography
SPIO Gd
MRA
Embolic mass
Individual coils
Artefacts
Flow in neck
Flow in sac
Embolic mass
Individual coils
Artefacts
Flow in neck
Flow in sac
Excellent Excellent
Acceptable Acceptable
None None
Excellent Excellent
Acceptable Acceptable
Excellent Excellent
None None
None None
Excellent Acceptable
Excellent Acceptable
Table 4 summarises the qualitative scores for the visualisation of the embolic mass, visualisation of individual filaments, visualisation of residual flow and artefacts under angiography and MRA. All subjects within the embolisation groups had identical scores. Loss of signal was not observed when comparing the MR images obtained immediately post-treatment and those obtained at 13 weeks post-treatment. Images of experimental aneurysms embolised with hydrogel filaments containing gadolinium or SPIO are shown in Figs. 2 and 3, respectively. Figure 4 shows images of experimental aneurysms embolised with platinum coils as a reference.
Angiography Aneurysms embolised with hydrogel filaments containing gadolinium or SPIO had equivalent angiographic results. The embolic mass and individual filaments were readily visible under angiography (Figs. 2a, b and 3a, b). The visualisation of the flow inside the aneurysm neck was evident during angiography. Residual flow within the aneurysm sac was difficult to distinguish (Figs. 2a and 3a). Artefacts were not present.
Fig. 2 Hydrogel filaments containing gadolinium. a Post-treatment digital subtraction angiography showed near complete occlusion of the aneurysm (red circle) treated with hydrogel filaments loaded with gadolinium. b Post-treatment angiography of the aneurysm (red circle) treated with hydrogel filaments loaded with gadolinium. The hydrogel filaments are visible but are less radiopaque than platinum coils. c Transverse, T1-weighted image of the aneurysm (red circle) immediately post-treatment. Hyperintense blood flow was observed in the carotid and vertebral arteries. The embolic mass of the hyperintense hydrogel filaments was visible in the aneurysm sac. d Transverse, T2-weighted image of the aneurysm (red circle) immediately post-
treatment. Neither the blood flow nor the embolic mass of hydrogel filaments was visible in T2-weighted imaging. e Transverse, 3D TOF image of the aneurysm (red circle) immediately post-treatment. Hyperintense blood flow was observed in the carotid and vertebral arteries. The embolic mass of the hyperintense hydrogel filaments was visible in the aneurysm sac. f Maximum intensity projection reconstruction of the aneurysm (red circle) and surrounding vasculature immediately post-treatment. Both the blood flow and embolic mass of hydrogel filaments are hyperintense, confounding the ability to distinguish between the two
MRA The embolic mass of hydrogel filaments loaded with gadolinium was hyperintense in T1-weighted imaging (Fig. 2c), TOF imaging (Fig. 2e) and the MIP reconstruc-
454
Neuroradiology (2011) 53:449–456
Fig. 3 Hydrogel filaments containing superparamagnetic iron oxide. a Three-month follow-up digital subtraction angiography showed complete occlusion of the aneurysm (red circle) treated with hydrogel filaments loaded with SPIO. b Three-month follow-up angiography of the aneurysm (red circle) treated with hydrogel filaments loaded with SPIO. The hydrogel filaments are visible but are less radiopaque than platinum coils. c Transverse, T1-weighted image of the aneurysm (red circle) at the 3-month follow-up. Hyperintense blood flow was observed in the carotid and vertebral arteries. The embolic mass of the hypointense hydrogel filaments was visible in the aneurysm sac. d
Transverse, T2-weighted image of the aneurysm (red circle) immediately at the 3-month follow-up. The blood flow was not visible. The embolic mass of hydrogel filaments was only faintly visible. e–g Sequential transverse, 3D TOF images of the aneurysm and parent artery (red circle) at the 3-month follow-up. Hyperintense blood flow was observed in the carotid and vertebral arteries. The embolic mass of the hypointense hydrogel filaments was visible in the aneurysm sac. h Maximum intensity projection reconstruction of the aneurysm (red circle) and surrounding vasculature at the 3-month follow-up. The blood flow was hyperintense and the embolic mass was hypointense
tion (Fig. 2f). The embolic mass inside the aneurysm sac and flow in the carotid and vertebral arteries were readily observable. However, due to their similar hyperintense appearance, it was not possible to distinguish between perfusion of the aneurysm sac and the hydrogel filaments. Similar to the in vitro results, the gadolinium-loaded
filaments were difficult to distinguish on T2-weighted imaging (Fig. 2d). Despite visibility of the embolic mass of hydrogel filaments, individual filaments could not be identified. Artefacts were not present. The SPIO-containing filaments were hypointense in T1weighted (Fig. 3c), T2-weighted (Fig. 3d) and TOF
Fig. 4 Platinum coils. a Post-treatment angiography showed near complete occlusion of the aneurysm (red circle) treated with platinum coils. b Transverse, TOF image of the aneurysm (red circle) immediately post-treatment. Hyperintense flow was observed in the carotid and vertebral arteries. Hypointense platinum coils are visible in
the aneurysm sac. c Maximum intensity projection reconstruction of the aneurysm (red circle) and surrounding vasculature immediately post-treatment. Like the DSA, filling in the aneurysm neck was observed
Neuroradiology (2011) 53:449–456
imaging (Fig. 3e), respectively. However, the SPIO-loaded filaments were more difficult to distinguish on T2-weighted imaging (Fig. 3d). In the MIP reconstruction (Fig. 3f), the blood flow was readily apparent, but the SPIO-loaded filaments were only faintly visible. The embolic mass of SPIO-loaded filaments was readily apparent; however, individual filaments could not be identified. Artefacts were not present.
Discussion In this study, we identified optimum concentrations of gadolinium and SPIO loaded into hydrogel filaments deliverable through microcatheters that resulted in visibility when imaged with MR. Hydrogel filaments loaded with barium sulphate and either gadolinium or SPIO were used to embolise experimental aneurysms. The embolic masses were visible in MRA immediately post-treatment as well as at the 13-week follow-up; however, additional work is required to develop hydrogel filaments that can be visualised individually. For the in vitro portion of the study, we evaluated gadolinium and superparamagnetic iron concentrations ranging from 1,500 to 7,500 and 500 to 2,500 ppm, respectively. From these results, we selected loadings of 4,500 ppm gadolinium and 2,000 ppm superparamagnetic iron for the in vivo studies. These values were selected because lower loadings resulted in inferior visibility, and higher loadings did not result in increased visibility. Signal artefacts were not present at any of the concentrations. For the Gd loadings, we used values that were significantly lower than those used in previous experimentation. Previously, microspheres with Gd loadings of 17,000 ppm have been prepared and evaluated [19]. One reason for the lower Gd concentration in this study is that we incorporated the Gd-DTPA into the polymeric backbone of the hydrogel filament, while the microspheres were prepared with Gd-DTPA entrapped in the polymer. The diffusion of Gd-DTPA out of the microspheres required higher initial loadings. For the SPIO loadings, we used values that were slightly higher than those used in previous experimentation. Previously, calibrated microspheres loaded with 670 ppm superparamagnetic iron were utilised for the embolisation of porcine kidneys [20] and a rabbit model of liver cancer [21]. Unlike in previous reports, we did not observe ferromagnetic distortions at superparamagnetic iron loadings of 1,100 ppm. The suitability of the Gd and SPIO loadings identified in vitro were confirmed in the experimental aneurysm model. The embolic masses of the Gd and SPIO-loaded hydrogel
455
filaments were visible during MRA. Furthermore, the visibility was maintained over the 13-week follow-up period. We believe that this durability is a result of the covalent incorporation of Gd-DTPA into hydrogel and physical entrapment of the 10-μm SPIO particles within the hydrogel. As expected, the hydrogel filaments loaded with gadolinium imaged hyperintense and those loaded with SPIO imaged hypointense. Our preference was for the SPIO-loaded hydrogel filaments as their hypointensity provided suitable contrast against the hyperintense blood flow. Together with the advances in MR hardware, MR sequences and MR-compatible microcatheters/wires, the hydrogel filaments described in this study represent a contribution in the progression towards real-time MRguided interventions. While challenges remain, some of the limitations of platinum coils have been overcome with the hydrogel filaments, including elimination of artefacts and visibility independent of blood flow. This study, although encouraging for the development of embolic devices with suitable characteristics for imaging with angiography and MRA, had limitations. First, the number of rabbits in each treatment group was small. This limitation is partially offset by the concordance of the results within the groups. Increasing the size of the groups would not alter the key points of this work. Second, the hydrogel filaments utilised in this study were prototype devices and were delivered via injection. The hydrogel filaments must be deliverable using a detachable pusher to have widespread clinical use in the treatment of intracranial aneurysms. Third, this study utilised MR-imaging sequences optimized for the human body. As the volume of rabbits is considerably smaller than that of humans, the image quality was poor. We are planning on addressing this limitation as we develop MR-imaging sequences for neurointerventional procedures. Having determined methods and loading concentrations for both gadolinium and SPIO, we are focusing on developing appropriate MR-imaging sequences for neurointervention and improving the handling characteristics of the hydrogel filaments, improving the visibility of single loops of hydrogel filaments and the development of a MR-compatible pusher and detachment system. Acknowledgements The authors gratefully acknowledge funding by MicroVention Terumo, the contributions of Drs. T. Hauser, R. Agic and M. Kral in creating the aneurysms and the contributions of M. Schober and B. Kinne in conducting the MRA. Conflict of interest statement E.K. and G.C. are employees of MicroVention Terumo. This study was funded by MicroVention Terumo, Aliso Viejo, CA.
456
Neuroradiology (2011) 53:449–456
References 1. Anzalone N, Scomazzoni F, Cirillo M, Righi C, Simionato F, Cadioli M, Iadanza A, Kirchin MA, Scotti G (2008) Follow-up of coiled cerebral aneurysms at 3T: comparison of 3D time-of-flight MR angiography and contrast-enhanced MR angiography. AJNR Am J Neuroradiol 29:1530–1536 2. Sprengers ME, van Rooij WJ, Sluzewski M, Rinkel GJ, Velthuis BK, de Kort GA, Majoie CB (2008) MR angiography follow-up 5 years after coiling: frequency of new aneurysms and enlargement of untreated aneurysms. AJNR Am J Neuroradiol 30:303–307 3. Saborowski O, Saeed M (2007) An overview on the advances in cardiovascular interventional MR imaging. Magma 20:117–127 4. Strother CM, Unal O, Frayne R, Turk A, Omary R, Korosec FR, Mistretta CA (2000) Endovascular treatment of experimental canine aneurysms: feasibility with MR imaging guidance. Radiology 215:516–519 5. Yang X, Bolster BD Jr, Kraitchman DL, Atalar E (1998) Intravascular MR-monitored balloon angioplasty: an in vivo feasibility study. J Vasc Interv Radiol 9:953–959 6. Elgort DR, Hillenbrand CM, Zhang S, Wong EY, Rafie S, Lewin JS, Duerk JL (2006) Image-guided and -monitored renal artery stenting using only MRI. J Magn Reson Imaging 23:619–627 7. Krombach GA, Wehner M, Perez-Bouza A, Kaimann L, Kinzel S, Plum T, Schibur D, Friebe M, Gunther RW, Hohl C (2008) Magnetic resonance-guided angioplasty with delivery of contrastmedia doped solutions to the vessel wall: an experimental study in swine. Invest Radiol 43:530–537 8. Ratnayaka K, Faranesh AZ, Guttman MA, Kocaturk O, Saikus CE, Lederman RJ (2008) Interventional cardiovascular magnetic resonance: still tantalizing. J Cardiovasc Magn Reson 10:62 9. Unal O, Li J, Cheng W, Yu H, Strother CM (2006) MR-visible coatings for endovascular device visualization. J Magn Reson Imaging 23:763–769 10. Kuehne T, Weiss S, Brinkert F, Weil J, Yilmaz S, Schmitt B, Ewert P, Lange P, Gutberlet M (2004) Catheter visualization with resonant markers at MR imaging-guided deployment of endovascular stents in swine. Radiology 233:774–780 11. Derdeyn CP, Graves VB, Turski PA, Masaryk AM, Strother CM (1997) MR angiography of saccular aneurysms after treatment
12.
13.
14.
15.
16.
17.
18. 19.
20.
21.
with Guglielmi detachable coils: preliminary experience. AJNR Am J Neuroradiol 18:279–286 Kang HS, Moon WJ, Roh HG, Han MH, Choe WJ, Cho J, Moon CT, Koh YC (2008) MR angiographic evaluation is limited in intracranial aneurysms embolized with Nexus coils. Neuroradiology 50:171–178 Hartman J, Nguyen T, Larsen D, Teitelbaum GP (1997) MR artifacts, heat production, and ferromagnetism of Guglielmi detachable coils. AJNR Am J Neuroradiol 18:497–501 Killer M, Kallmes DF, McCoy MR, Ding YH, Shum JC, Cruise GM (2009) Angiographic and histologic comparison of experimental aneurysms embolized with hydrogel filaments. AJNR Am J Neuroradiol 30:1488–1495 McCoy MR, Cruise GM, Killer M (2010) Angiographic and artefact-free computed tomography imaging of experimental aneurysms embolised with hydrogel filaments. Eur Radiol 20:870–876 Constant MJ, Keeley EM, Cruise GM (2009) Preparation, characterization, and evaluation of radiopaque hydrogel filaments for endovascular embolization. J Biomed Mater Res B Appl Biomater 89:306–313 Forrest MD, O'Reilly GV (1989) Production of experimental aneurysms at a surgically created arterial bifurcation. AJNR Am J Neuroradiol 10:400–402 Roy D, Milot G, Raymond J (2001) Endovascular treatment of unruptured aneurysms. Stroke 32:1998–2004 Faranesh AZ, Nastley MT, Perez de la Cruz C, Haller MF, Laquerriere P, Leong KW, McVeigh ER (2004) In vitro release of vascular endothelial growth factor from gadolinium-doped biodegradable microspheres. Magn Reson Med 51:1265–1271 Namur J, Chapot R, Pelage JP, Wassef M, Langevin F, Labarre D, Laurent A (2007) MR imaging detection of superparamagnetic iron oxide loaded tris-acryl embolization microspheres. J Vasc Interv Radiol 18:1287–1295 Lee KH, Liapi E, Vossen JA, Buijs M, Ventura VP, Georgiades C, Hong K, Kamel I, Torbenson MS, Geschwind JF (2008) Distribution of iron oxide-containing embosphere particles after transcatheter arterial embolization in an animal model of liver cancer: evaluation with MR imaging and implication for therapy. J Vasc Interv Radiol 19:1490–1496
