|Home | About | Journals | Submit | Contact Us | Français|
Aortic valves have been implanted on self-expanding (SE) and balloon-expandable (BE) stents minimally invasively. We have demonstrated the advantages of transapical aortic valve implantation (tAVI) under real-time magnetic resonance imaging (rtMRI) guidance. Whether there are different advantages to SE or BE stents is unknown. We report rtMRI guided tAVI in a porcine model using both SE and BE stents, and compare the differences between the stents.
Twenty-two Yucatan pigs (45-57kgs.) underwent tAVI. Commercially available stentless bioprostheses (21-25mm) were mounted on either BE platinum-iridium stents or SE nitinol stents. rtMRI guidance was employed as the intraoperative imaging. Markers on both types of stents were used to enhance the visualization in rtMRI. Pigs were allowed to survive and had follow-up MRI scans and echocardiography at 1, 3 and 6 months postoperatively.
rtMRI provided excellent visualization of the aortic valve implantation mounted on both stent types. The implantation times were shorter with the SE stents (60±14 seconds) than BE (74±18s), (p=0.027). Total procedure time was 31 and 37 minutes respectively (p=0.12). It was considerably easier to manipulate the SE stent during deployment without hemodynamic compromise. This was not always the case with the BE stent and its placement occasionally resulted in coronary obstruction and death. Long-term results demonstrated stability of the implants with preservation of myocardial perfusion and function over time for both stents.
SE stents were easier to position and deploy thus leading to fewer complications during tAVI. Future optimization of SE stent design should improve clinical results.
To reduce trauma and speed recovery for the patient; minimally invasive aortic valve replacement, both a percutaneous transfemoral approach [1-6] and a left ventricular transapical approach [7-12], have been reported. Compared to the former, transapical aortic valve implantation provides a direct and short access to the native valve. This approach may be more applicable to a number of patients because of the lack of physical anatomic limitations that can preclude a femoral approach.
MRI provides better anatomic detail than fluoroscopic or echocardiographic imaging without additional risk of radiation and contrast reaction. Calcification does not interfere with the MR imaging. Vascular as well as soft tissue visualization can easily be performed simultaneously. The development of real-time magnetic resonance imaging (rtMRI) allows this imaging modality to provide intraoperative guidance for delivery of prosthetic aortic valves . MRI also provides the ability to assess results, such as ventricular and valvular function, and myocardial perfusion, immediately after intervention. Our group has successfully performed transapical aortic valve replacements using rtMRI guidance [14, 15]. After acute feasibility studies confirmed the safety of the rtMRI guided procedure, initially using a BE stent, we commenced a series of experiments with long term follow up .
In minimally invasive aortic valve replacement, a stent is employed to support the prosthesis and to anchor the prosthesis to the aorta. Two types of stents, balloon-expandable (BE) and self-expanding (SE), have been used. One of the first percutaneous prosthesis for the treatment of aortic valve disease was the Cribier-Edwards Aortic PHV (Edwards Lifesciences Inc, Irvine, California). This balloon-expandable prosthesis is a tubular slotted stainless steel stent with an attached equine or bovine pericardial trileaflet valve [5,6,16-17]. This prosthesis has also been used for both a transfemoral and a transapical approach [8-11]. The other percutaneous prosthesis which has been used in a large clinical trial is the self-expanding CoreValve ReValving prosthesis, (CoreValve Inc, Irvine, California). This prosthesis uses a porcine bioprosthesis within a nitinol frame [18-20].
There are several other new stented prostheses in various stages of development for percutaneous approaches [7,21-23]. While there are many advantages and perhaps several disadvantages with all of these devices, whether there are additional benefits or risks depending on the mechanism whereby the stent expands and the valve is subsequently deployed are unknown.
In this paper, we report our rtMRI guided tAVI in a porcine model using both SE and BE stents, and compare the differences between the stents, in terms of the ease of performing the tAVI, the valve function, the valve position and stability as well as stent integration with the aorta in a preclinical survival experiment.
As the name implies, the SE stent can expand without external force and can be crimped before it is allowed to expand. For our study we designed a stent that could be used to accommodate a conventional stentless aortic porcine bioprosthesis (Toronto SPV, St. Jude Medical, Inc, Minneapolis, MN or Freestyle, Medtronic Inc., Minneapolis, MN), whose clinical durability and success rate are already been proven. As the aortic valve lies in close proximity to both the anterior leaflet of the mitral valve and the coronary ostia, the correct position and orientation of the implanted valve is critical. Misalignment of the prosthesis could result in mitral valve damage or cardiac ischemia. The prosthesis must be adequately fixed in place such that migration or embolization does not occur despite high blood pressures.
The design and development of the SE stent addressed the above requirements. The stent is made from nitinol; it is safe and compatible in the high magnetic field of MRI. The material is memory alloy. The stent can be crimped, and it will return to its original shape when there is no external force. It was created with different diameters to accommodate the aforementioned 21-25mm stentless bioprosthetic valves that are used clinically to match the patients’ aortic annulus. A special cylindrical geometric design with flare at both ends was employed to ensure better fixation and scaffolding. All the values of the geometric parameters guarantee even radial force and flexibility as well as rigidity of the stent.
The structure of the stent also makes it easily retractable into a delivery device. Adjustment of the position during valve placement is therefore possible. Figure 1 left shows a St. Jude Toronto SPV valve (St. Jude Medical Inc, Minneapolis, MN) mounted on the SE nitinol stent. Small austenitic stainless steel fragments (0.5mm) were welded on the side of the stent aligned with one of the prosthesis commissures. This paramagnetic passive marker is visible as a dark signal in the MRI and is used to indicate the orientation of the stented prosthesis.
A commercially available platinum-iridium stent (Cheatham Platinum, NuMed, Hopkinton, NY) was used for BE stented prosthesis. The BE stent was originally developed, approved and commercially available for treatment of coarctation of the aorta. It is MR safe and compatible. 21-25mm stentless bioprostheses (Toronto SPV, St. Jude Medical, Minneapolis, MN, or Freestyle, Medtronic Inc, Minneapolis, MN) with proper size based on pre-op MRI scan were mounted on the stent. Figure 1 shows a Freestyle (Medtronic Inc, Minneapolis, MN) mounted on the Numed platinum-iridium stent. The same passive markers were also welded to these stents. The stented prosthesis is then circumferentially compressed over two-stage balloon-tipped catheters (NuMed, 25-30mm OD, 50mm long). A central guide wire allowed passage and tracking of the balloon-tipped catheter antegrade across the native aortic valve.
1.5-T Magnetom Espree (Siemens Medical Solutions, Munich, Germany) was used for the intervention and post intervention assessment. A real-time Interactive MRI system connected to the MRI scanner provides visualization of the progress of the procedure . This system is comprised of an interactive user interface, an operating room large-screen display, gated pulse sequences, and image reconstruction software. High-quality images can be obtained at 5-10 frames per second with low latency. With the real-time interactive MRI, multiple oblique slices can be obtained in rapid succession and can be simultaneously rendered in their relative 3D positions to provide optimal 3D anatomic information. Image contrast, image plane orientations, acquisition speed, 3D rendering, and device tracking can be readily adjusted as needed during scanning.
An active wire (ie, MR signal receiving loopless antennae) is used to indicate the orientation of the stented prosthesis. This wire was colorized and highlighted in the MR images. For the BE stent, the wire is temporarily affixed to the stent/catheter, and aligned with one of the valve commissures . For the SE stent, it is embedded in the delivery device, and is aligned with one of the valve commissures loaded in the delivery device.
All experiments were performed by a single operator under protocols approved by the National Institutes of Health Animal Care and Use Committee. 1.5T MR imaging was used to precisely identify the anatomic landmarks of the aortic annulus, coronary artery ostia, and the mitral valve leaflets. Based on the pre-op image, an appropriate size prosthesis was selected and mounted on either a BE or a SE stent of matching size. The animals then underwent rtMRI guided transapical aortic valve replacement without unloading by rapid ventricular pacing or cardiopulmonary bypass. Details of the real time MR imaging during transapical aortic valve implantation have previously been reported [14, 15].
To date we have operated on 42 Yucatan pigs, 20 for an acute feasibility study and the remaining 22 for a survival procedure comparing SE and BE stents. All the animals for acute feasibility study received BE prostheses. For the long term survival study, 11 animals received a BE prosthesis and 11 animals had implantation with SE prosthesis. We recorded the time of the procedures and observed the interference of the devices with the surrounding environment during the procedures. Post-placement images were acquired to evaluate the prostheses position, as well as, the valvular and heart function.
The procedures using the two types of the stents are slightly different. In a procedure using the BE prosthesis, the distal end of the sheath of the loaded delivery device is placed below the aorta annulus level; pushing the balloon tipped catheter will advance the crimped prosthesis out of the sheath to the desired position, and then the balloon was inflated to expand the stent and deploy the prosthesis. In a procedure using the SE prosthesis, the loaded delivery device was first advanced into the ascending aorta. The retraction of the sheath allowed crimped prosthesis to expand and affix to the desired position. Figure 2 and figure 3 shows the progress of the BE prosthesis and SE prosthesis valve implantation under MRI guidance, respectively. We recorded the time for the procedures and compared the ease of performing the procedures with both stents.
During the procedure, the animals were heparinized and monitored with ECG, oxygen saturation, end-tidal CO2, systemic and left ventricular blood pressure, and arterial blood gas analysis.
At 1 and 3 months postoperatively, follow-up MRI scans and transthoracic echocardiograms were acquired while at 6 months postoperatively MRI scans and confirmatory 2- and 3-dimensional transesophageal echocardiograms were acquired. After 6 months the animals were sacrificed, and the histopathology analyses were performed. We compared the valvular function, the valve position and stability, stent integration with the aorta for both stents during the long-term follow up.
rtMRI provided excellent visualization of the aortic prosthesis implantation mounted on both stent types. The active wires were a superb indicator of the valve orientation in MRI. The passive markers on the SE stents were also help to identify the valve orientation. These markers were somewhat difficult to visualize via MRI when the stents were fully crimped but became more apparent as the stents were deployed.
We recorded the procedure time and deployment time for using BE stent and SE stent on 22 survival animals. We also performed t-test to determine the significance of the differences on the procedure time and deployment time between two groups. Total procedure time was 37 and 31 minutes for using BE stent and SE stent respectively. The total procedure time for both stents were not significantly different (p=0.12). The time from introduction of the prosthesis into the trocar to deployment the stent is fully expanded (deployment time) was 74±18 seconds (mean ± std. dev.) and 60±14 seconds (mean ± std. dev.) respectively. This deployment time was significantly shorter for the SE stent (p=0.027). Once the BE prosthesis was in place, the balloon is first partially inflated by using normal saline mixed 100:1 with an MR contrast agent Gd-DTPA (Magnavist, Berlex Inc, Montville, NJ), the position is re-confirmed to be ideal and the balloon is then fully expanded and the stented prosthesis completely deployed. The procedures using BE prosthesis therefore took a slightly longer time because of this time for staging the balloon-inflation and the difficulty in orienting the valve knowing that once the balloon was completely inflated there was no margin to allow for adjustment. When the balloon was inflated to deploy the BE stent there was a significant change in the blood pressure and a concomitant decrease in the cardiac output. This hemodynamic compromise was seen even with partial inflation of the balloon. There was no significant blood pressure drop observed during the SE prosthesis implantation. Even partially deployed the SE stent was not as obstructive as the combination of balloon and stent for the BE device. Blood can flow through a partially deployed SE stent but even a partially filled balloon for a BE stent is obstructive and had a negative impact on the hemodynamic stability of the animal. Additionally the inability to easily control deployment with the BE stent resulted in coronary obstruction and death. 4 out of 20 of the animals in acute feasibility study with BE prosthesis implanted died of coronary obstruction.
Post-placement images were acquired to confirm the positions of the prostheses and assess the valvular and heart function. Gated cine MRI revealed excellent myocardial function after valve implantation in both long- and short-axis views for animals in whom the valves were appropriately positioned. The phase-contrast CINE MR images confirmed good systolic flow with excellent valve leaflet opening and no evidence of turbulence, diastolic regurgitant flow, or paravalvular leak. First-pass perfusion studies  demonstrated adequacy of myocardial blood flow after valve placement in all animals following successful deployment. The perfusion results confirmed adequacy of blood flow at the tissue level, indicating proper valve positioning with respect to the coronary ostia.
Among 11 animals implanted with a BE prosthesis, 3 died between 1 and 2 weeks postoperatively of respiratory complications. Among 11 animals implanted with SE prosthesis, 3 animals died within one month postoperatively due to late pneumothorax or pneumonia, 1 died of pneumonia at day 35. None of the survivors of the initial implantation died of late coronary obstruction. Hemodynamic measurements of ventricular and valvular parameters of 15 animals based on 6 months follow-up echocardiograms (and confirmed by MRI scans) are presented in figure 4. There is no significant difference between the animals with the BE prosthesis and those with the SE prosthesis. The degree of insufficiency of the aortic valve and the mitral valve at 6 months after the procedure were measured and are shown in figure 5. More aortic and mitral regurgitation was seen with the BE prosthesis but this trend was not statistically significant.
Post-mortem pathologic analysis, after sacrifice at 6 months, verified that the implanted both BE and SE prosthesis appeared in place in the aortic root. The prosthetic commissures were incorporated with neointimal growth continuous with the native leaflet commissures. The average strut fractures for the platinum iridium BE stent was 5.0±3.1 (mean ± std. dev.), while the average fractures for the nitinol SE stent was 1.6±2.5 (mean ± std. dev.)(p=0.046). There was no particular pattern of strut fractures observed. In eight animals with BE stent, 2 animals had fractures at the level of the proximal crowns; 3 animals had fractures at the level of the distal crowns, and 3 animals had fractures scattered throughout the stent. In seven animals with SE stent, 1 animal had fracture at the level of the proximal crowns, 1 animal had fractures at the level of distal crowns, and 2 animals had fractures throughout the stent, 3 animals did not have fractures. The fractures are due to the stent material fatigue and the expansion, contraction, torsion between the aorta and the stent. We performed a t-test to determine the significance of the differences in the number of strut fractures between the two groups. The nitinol SE stent has elasticity and its geometry design permit to handle the torsion better, while the BE stent has no elasticity and the material is relatively soft.
Some limitations were only observed in BE stent cases at the time of sacrifice. In one animal, the stent was collapsed/compressed on itself at its mid-section. Malapposed stent struts at either ends were observed in five animals. In one animal, a prosthesis cusp appeared collapsed. All of these were due to uneven deployment of the BE prosthesis. On the other hand in the SE cases, transverse sections from the device (prosthetic aortic valve) showed a widely patent implant lumen with circular, symmetrical shape maintaining tight apposition of the stent frame to the aortic wall in all the animals with the SE prosthesis. Representative radiographs of the BE stented prosthesis and the SE stented prosthesis are shown in figure 6. Autopsy confirmation of BE and SE prosthetic valve location after 6 months implantation is seen in figure 7.
rtMRI provides excellent visualization for intraoperative guidance of aortic valve replacement on the beating heart using both BE and SE stented prosthesis.
The uncertainty of the stented prosthesis’ orientation and position during the expanding of a balloon, may lead to the blockage of the coronary ostia. This was observed in our previous acute studies. Since the stented prosthesis is not rigidly mounted on the balloon, there is the possibility that the prosthesis slides and rotates on the balloon during the balloon-expansion. Moreover, the balloon may not expand uniformly, causing the prosthesis to move from its position or be unevenly deployed. As a result, coronary cusp may collapse and focal gaps between stent and the native aortic wall may occur. The difficulty with uneven deployment is due in part to the uneven expansion of the balloon when it is inflated. The force of the balloon is not equally distributed to the stent. This may be an even greater problem in calcified aortic valves where the calcium will impair balloon filling and stent expansion. When the radial force and position of the stent is primarily due to the balloon then uneven expansion leads to uneven force and unbalanced strut expansion and would lead to long term unbalanced distribution of force on the stent with expansion and contraction and torsion of the aorta.
To deploy a BE stent, additional assistance may be required to inflate the balloon while the surgeon controls the positioning. The balloon needs to be gradually inflated so that the orientation and position of the prosthesis can be adjusted before it is fully affixed. The more the balloon is inflated the more obstructive it is to the left ventricular outflow track. To avoid significant hemodynamic compromise or embolization of a partially deployed device, unloading the heart by cardiopulmonary bypass or minimizing ejection by rapid ventricular pacing has been used clinically to assist in the placement of BE stented prostheses. The BE stent is not retrievable. Once it is outside of the protecting sheath, it cannot be easily retracted. One advantage of the BE stent is its “universal” size. The final expanded size of the stent depends in part on the size of inflated balloon and can therefore be used to allow the stent and prosthesis to conform to the native valve and annulus.
In our experience, SE stents were easier to position and deploy thus leading to fewer complications during transapical aortic valve implantation. The intrinsic radial force allows for even expansion of the prosthesis. As a result, the orientation of the implanted valve is more predictable. The elimination of the balloon inflation avoids the possibility of the hemodynamic compromise. Partially deployed the SE stent allows blood flow through the stent and there is no balloon to impede flow. The SE stent can be retrieved and repositioned before it is fully expanded; this helps with precise placement and diminishes the risk and embolization. All of these features favored the SE stent. This ease of implantation may be a subjective finding with any device but is nevertheless important to the practioner.
Additionally the long term impact of balloon expansion on the prosthetic valve leaflets is unknown. For both stents the valves are similarly crimped but only the BE stent puts additional stress on the leaflets of the prosthesis as the balloon is inflated and stretches the leaflets as well as the stent. This could lead to short and long term damage and lessen durability.
Although we performed our experiments under intra-operative rtMRI guidance, the materials and devices presented in the paper can be used in the transapical aortic valve implantation under guidance with other imaging modalities. The conclusions are independent to the image guidance method and the results apply to those centers without access to intra-operative rtMRI.
We continue to evaluate the long term results of using both stents. Further optimization of the SE stent design should facilitate the procedure and may improve clinical results. One of the earliest upgrades was the development of a stent retrieving device, which can be used to retrieve and reposition the stent as well as prevent unexpected motion of the stent during deployment.
The authors also would like to thank Dr. Renu Virmani and Russ Jones from CV Path Institute, Inc for the pathology analysis.
The authors are supported through the Intramural Research Program of the National Heart, Lung, and Blood Institute, NIH, DHHS.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Keith A. Horvath, Cardiothoracic Surgery Research Program, National Heart, Lung and Blood Institute, National Institutes of Health.
Dumitru Mazilu, Cardiothoracic Surgery Research Program, National Heart, Lung and Blood Institute, National Institutes of Health.
Ozgur Kocaturk, Translational Medicine Branch, National Heart, Lung and Blood Institute, National Institutes of Health.
Ming Li, Cardiothoracic Surgery Research Program, National Heart, Lung and Blood Institute, National Institutes of Health.