| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Eur J Cardiothorac Surg 2006;29:703-708
© 2006 Elsevier Science NL
a Universität Leipzig, Herzzentrum, Klinik für Herzchirurgie, Strümpellstrasse 39, 04289 Leipzig, Germany
b Cardiopulmonary Research Science and Technology Institute, Department of Cardiac Surgery, Dallas, TX, USA
c Klinik für Thorax-, Herz- und Thorakale Gefässchirurgie, JW-Goethe Universität Frankfurt, Germany
d Universität Leipzig, Herzzentrum, Klinik für Kardiologie, Leipzig, Germany
Received 26 October 2005; received in revised form 16 January 2006; accepted 30 January 2006.
* Corresponding author. Tel.: +49 341 865 1424; fax: +49 341 865 1452. (Email: walt{at}medizin.uni-leipzig.de).
 |
Abstract |
|---|
 Top Abstract 1. Introduction2. Materials and methods3. Results4. Discussion5. ConclusionAppendix AReferences |
|---|
Key Words: Aortic valve replacement • Stent-fixed aortic valve • Transapical approach • Beating heart cardiac surgery
|
1. Introduction |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. Discussion5. ConclusionAppendix AReferences |
|---|
In 2003, approximately 10,000 patients received isolated AVR in Germany. In-hospital mortality was 2.4% after mechanical and 3.8% after xenograft implantation [2]. Subgroups, especially octogenarians have a higher risk of up to 10% during aortic valve surgery. According to the STS database, the operative mortality after AVR was 4% in a total of 32,968 patients operated in 2004 in the United States. Concomitant coronary artery bypass graft (CABG) surgery lead to further increase in mortality up to 6.8%, 6.6% in patients with diabetes, 7.6% with previous cardiac surgery, and 17% in patients with end stage renal disease. These data are a clear indicator for a significant increase in perioperative risk in presence of patient-related comorbidites.
Application of the numeric and logistic EuroSCORE allows the preoperative estimate of perioperative risk [3,4]. A subgroup of patients with aortic valve disease carries a risk of 15% or above and some have even been considered nonsurgical candidates. In this high risk group percutaneous aortic valve implantation has been performed as a treatment option [5–7]. In order to further reduce the invasiveness of standard surgical AVR, we evaluated a facilitated transapical antegrade approach for potential use in high-risk surgical patients.
|
2. Materials and methods |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. Discussion5. ConclusionAppendix AReferences |
|---|
Anesthetic induction was per standard techniques, using ketamine (2.2 mg/kg), xylazine (1.1 mg/kg), and tiletamine/zolazepam (Telazol®) (4.4 mg/kg) IM (IR) or atropine (0.02 mg/kg) azaperone (Stresnil®) (2 mg/kg) IM plus intravenous thiopental (Trapanal®) (4 mg/kg) and fentanyl (LE) for induction. This was followed by endotracheal intubation, controlled ventilation using isoflurane and intravenous access. All procedures were performed in the dorsal recumbent position.
2.1 Technical equipment
Both laboratories were equipped to perform routine cardiac surgical operations including a standard cardiopulmonary bypass (CPB) system; access was obtained via the femoral vessels. ECG, arterial blood pressure, and central venous pressures were routinely monitored. Functional hemodynamic monitoring was performed using intracardiac ultrasound (IR) with the probe advanced via the right internal jugular vein or standard transesophageal or epicardial echocardiography (LE).
Additional equipment to perform fluoroscopic examinations was available in both laboratories. This consisted of a mobile C-arm (GE Stenoscop 9000+) allowing for conventional monoplane visualization (IR). In the other laboratory (LE) a monoplane fluoroscopic angiography system (Axiom Sensis, Siemens, Munich, Germany) was used. Fluoroscopy should ideally be performed to give a perpendicular view of the aortic root. This allows for optimal delineation of the level of the aortic annulus in relation to the aortic sinuses together with imaging of the coronary ostia. After testing multiple angulations of the system, the preferred visualization of the aortic root was achieved at a left anterior oblique 35° and cranial 15° position of the system.
2.2 Sutureless stent-fixed aortic valve
A 23-mm Cribier–Edwards equine pericardial valve (model 9000) mounted on a stainless steel stent was used. The valve has three cusps and three commissures as is seen with other pericardial xenografts. The stainless steel stent has a very low profile when fully expanded. The lower inflow portion of the valve is covered with polyethylene terephthalate (PET) cloth. This is the same device as used in recent clinical percutaneous approaches [4–6]. One size was available for implantation fitting all annular sizes up to 23 mm (Fig. 1
). The sutureless stent-fixed aortic valve was prepared for antegrade delivery under sterile conditions in the operating room by a technician just prior to implantation. The delivery catheter was flushed with a heparinized saline solution. The deployment balloon was primed with a mixture of saline and contrast, free of air. A balloon-sizing ring was used to determine the exact amount of volume needed to expand the balloon to 22 mm. The valve was crimped onto the deployment balloon so that it was equidistant between two radiopaque markers and was able to be passed through the 24-F delivery sheath; this was performed immediately prior to delivery. All valve deployments were performed using standard volumetric inflation of the balloon.
|
A surgical cut down was performed to cannulate the right internal carotid artery for placement of a calibrated angiography catheter (pigtail). Echocardiography was performed to aid with annular sizing, implantation, and evaluation of valve performance. After opening the pericardium at the apex, a double purse string suture of 3-0 Prolene was placed. Anticoagulation was performed using 300 IU/kg of heparin. Cineangiography of the aortic root with a calibrated catheter was then performed to size the aortic annulus; landmarks were used to identify the noncalcified porcine annulus.
A 035'' super-stiff guidewire was placed transapically and across the aortic valve down the descending thoracic aorta using the Seldinger technique. Balloon dilatation of the aortic valve was performed in order to mimic the situation of clinical use in potential patients with aortic valve stenosis. The 24-F introducer sheath was then placed into the LV below the aortic valve under fluoroscopic guidance. The crimped valve on the delivery catheter (Fig. 2 ) was then introduced through the delivery sheath into the LV. After cardiac unloading the valve was positioned so that the annulus bisects the stent. Small contrast boluses were given to aid in identifying the noncalcified porcine annulus and the coronary ostia. Once positioned, the valve was deployed by inflating the delivery balloon with saline mixed with contrast to achieve full expansion of the stent. Control angiography confirmed exact valve position (Fig. 3 ). The heart was loaded and the guidewire and sheath retrieved, the apex was safely secured using the purse-string sutures. Valve function was assessed using cineangiography and echocardiography after eventually weaning off CPB. The animals were then sacrificed and pathological examination of the heart performed.
|
|
|
3. Results |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. Discussion5. ConclusionAppendix AReferences |
|---|
Eight procedures were performed at IR and seven procedures at LE. Mean body weight of the animals was 41.8 ± 3.2 kg. The mean aortic annulus diameter was 19.9 ± 1.4 mm on fluoroscopy and 20.2 ± 0.9 mm by echocardiographic measurements. This slight difference was statistically not significant. Thus, adequate measurements were obtained using either method. Similarly, there were no significant differences between the experiments performed in IR and LE. Detailed results on the parameters of the experimental model are given on Table 1 .
|
The prosthesis was initially positioned as targeted in all 15 evaluations. Fluoroscopic guidance was used as the primary method to guide implantation in all cases. In some cases, slightly higher or lower position was intentionally aimed at. Nine of the valves were in stable position and remained there after placement. The use of landmarks was helpful after preimplantation visualization of the aortic root. Distal embolization occurred in three cases, two of which having no ventricular unloading as previously described. Proximal embolization occurred in other three cases. In one animal embolization occurred after placing the prosthesis in a relatively low position to prevent occlusion of the left coronary ostium.
Postimplantation aortic incompetence was present in five cases. This was due to true paravalvular leakage in two and due to trans-stent backflow laterally directed in the other three animals. Coronary occlusion, partial or complete, occurred in three cases. All patients receiving on-pump valve implantation with the valve in the right position were weaned off bypass. Echocardiography revealed good cusp motion in all cases with the valve in the right position. A summary of perioperative results is given in Table 2 .
|
|
4. Discussion |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. Discussion5. ConclusionAppendix AReferences |
|---|
4.1 Literature results
Percutaneous AVR has been performed in selected patients [5–7]. Besides successful implantations, problems have arisen with regard to trans-septal puncture and the rather long distance to the aortic valve, impairing the positional stability of the prosthesis during implantation [6,8]. Furthermore, femoral access is problematic in patients with peripheral arterial occlusive disease. All patients included into clinical trials so far had been rejected by two independent cardiac surgeons [5–7]. There are reports identifying ‘nonsurgical candidates’, with some even talking about large cohorts of untreated patients. Depending on the individual clinician indications and attitudes to this disease, this may be rather surprising. Given the good results of conventional aortic valve surgery, therapy should not be denied to those suffering from symptomatic aortic valve disease. However, with the increasing number of individual patients with higher surgical risks, the search for less invasive techniques, the wish to be open minded towards new approaches and an attitude to adopt new techniques the transapical approach may be an option in future.
Different experimental approaches for direct aortic valve replacement have been published. Hubers group successfully implanted a prototype-valved stent in 12 pigs with transapical delivery [9]. Similar valves with nitinol Z-stents had been previously evaluated in the pulmonary and mitral position by the same group [10,11]. Percutaneous resection of calcified structures before valve implantation has been studied experimentally [12]. However, for implantation of the Cribier–Edwards prosthesis resection is not required. Good coronary flow has been proven for a self-expanding nitinol stent in the aortic position [13,14]. In comparison to that device, the steel stent-based Cribier–Edwards valve should not cover the coronary ostia and thus not lead to an impairment of coronary flow in presence of a normal (human) length of the aortic root. Bonhoeffers group deserves the credit for their large clinical experience with percutaneous valve placement, even though not in the aortic but in the pulmonary position [15].
4.2 Team approach
Traditionally AVR is a surgical task. The prostheses developed over the past 45 years and have reached a high level of perfection allowing patients to continue a rather healthy life with significant improvement in their quality of life [16,17]. When applying any new technology, an interdisciplinary research group with a wide variety of experiences, including cardiologists and cardiac surgeons, is helpful. Joint collaborative efforts and experiences will provide maximum benefit for potential patients.
4.3 Experimental model
The porcine model used is similar to human anatomy, but it differs in several critical features. Specifically the aortic annulus was pliable because annular and leaflet calcification was not present. Thus, anchoring of the prosthesis is not comparable to that which is potentially possible in the presence of equally distributed calcification as in a human clinical situation. In addition, the relationship of the coronary ostia to the annulus and to the mitral valve annulus was closer. No ideal animal model mimicking aortic valve stenosis together with adequate calcification is known.
Clinical applications of the prosthesis in the presence of calcified structures may be easier than in this experimental model as visualization will be better due to the calcium and fixation of the steel stent will be enhanced after pressing the calcium to the side. Calcification will, in addition, lead to reduced pliability of the aortic annulus and the aortic root and thus to better valve fixation. Another important aspect is the distance between the aortic annulus and the coronary ostia. This was rather short (7–8 mm) in the experimental model, leading to a higher risk of coronary artery obstruction after positioning of the prosthesis. However, in humans the geometry of the aortic root is different with a larger distance between 12 and 15 mm leading to a significantly lower risk of coronary ostia obstruction in potential clinical applications. Furthermore, better visualization in presence of some annular and ostial calcifications will be present. This study was performed solely to address the feasibility of the delivery technique and not the valve itself. Accurate placement of the same valve without migration has been documented by the percutaneous approach in humans. Thus, the migration of valve we encountered in our study, we feel, is a reflection of the animal model and not the valve or the delivery technique.
4.4 Sutureless stent-fixed aortic valve prosthesis
The Cribier–Edwards stent-fixed aortic valve prosthesis has been applied in the clinical setting using a percutaneous approach. This design will allow for sufficient hemodynamic performance as proven during experimental testing and from initial clinical results. Durability may be sufficient, especially in higher risk patients who may have a limited life expectancy. From the valve design standpoint, the stent is minimally covered with cloth and there is no cuff around the outer side. This may theoretically allow for optimal fixation in a calcified annulus. However, in presence of some irregularities at the annular level, typical paravalvular as well as trans-stent leakage may occur [4–8]. The latter would be directed from the central flow area through the stent depending of the position of the valve in relation to the native aortic root. Adding a cuff or skirt to the stent may reduce this risk. In the experimental model we observed aortic incompetence in several cases due to lack of backpressure. Further evaluation in the clinical setting is warranted. We strongly believe that it is important, whenever new prostheses are developed and applied, to aim at achieving the present high level of good clinical results with conventional devices. Minimal incompetence is acceptable in high-risk patients if a safer way to implant the prosthesis is available. However, a higher degree of valvular incompetence may not be tolerated by a ventricle used to aortic valve stenosis and should not be accepted.
4.5 Access for transapical AVR
Access to the apex of the heart was sufficiently obtained via a partial small inferior sternotomy in this model. Transposing these results into potential human practice and considering the different anatomies an anterolateral minithoracotomy in the fifth or sixth intercostal space would be feasible. This would even be true in the presence of cardiac adhesions during potential redo operations.
4.6 What is the ideal candidate for transapical AVR?
Transapical AVR is a procedure in its infancy. Early experimental data are promising and limited access beating heart aortic valve placement may become a reality in selected patients. However, there is no consensus yet as to who may be an ideal patient. Equally distributed calcification of the aortic annulus and the native valve cusps is important to safely position and anchor the valve. Larger eccentric plaques may lead to uneven opening of the valve or potential paravalvular leakage. Presence of bicuspid aortic valve disease or fusion of one of the commissures should be considered carefully. Transesophageal echocardiographic visualization is the standard for an exact preoperative diagnosis. In addition, CT scanning to delineate the amount of calcium in the aortic root may be helpful. Complete equal opening of the commissures without fusion with some annular and cusp calcification may be optimal. Patients with previous coronary artery bypass graft surgery with open grafts including a patent left internal mammary artery (LIMA) may be ideal candidates. Redo aortic valve surgery usually poses patients at an increased surgical risk due to advanced age and additional risk factors. Open bypass grafts would provide an additional safety margin in case of a short distance between the annulus and the coronary ostia. Other suitable candidates would be patients with porcelain aorta without any surgical option at an acceptably low perioperative risk. A stiff and calcified aortic root would be an additional help to safely anchor a sutureless stent-fixed aortic valve prosthesis under such circumstances. Patients carrying a higher surgical risk profile would be candidates to receive beating heart minimally invasive aortic valve placement with femoro-femoral CPB backup. Recent clinical experiences indicate that using an ‘oversized’ valve technique of approximately 3 mm will be important. Therefore, the 23 mm device should be used for patients with rather small aortic annuli of 19 mm to maximal 21 mm. If available, a larger 26-mm device could be considered for patients having an aortic annulus up to 24 mm.
|
5. Conclusion |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. Discussion5. ConclusionAppendix AReferences |
|---|
|
Appendix A |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. Discussion5. ConclusionAppendix AReferences |
|---|
Dr M. Krason’ (Zabrze, Poland): I would like to ask you what is your positioning, let's say tactics, and what is your oversizing tactics in this area of self-expanding stents, or balloon-expanding stents? Could you give us some more ideas, having in mind that there were so many displacements of the device?
Dr Walther : It is a very important question. We aim at some oversizing. In the animal model that is difficult, because the annulus is very flexible; it is noncalcified. I would expect that in humans with a calcified annulus there is more rigidity. So the anchoring of the valve would be easier. We measured the annulus in these animals on echo and on fluoroscopy, and we had an annulus of 19.8 mm on echo and 20.1 mm on fluoroscopy and we implanted a 23 mm valve. So we aim at something like 3 mm oversizing of the prosthesis.
Dr A. Böning (Kiel, Germany): Could you comment on the removal of the original aortic valve leaflets?
Dr Walther : Well, in the animal we didn’t remove them. In patients, we would really look for patients who have an equally distributed calcification, and we think that the calcification will help us to anchor the valve. As far as I heard from initial experiments in the 1990s from Heartport that they tried to do research on aortic valve implantation and tried to remove the whole calcium and they failed because they weren’t able to safely remove all that, and we think that the calcium may be a kind of anchor mechanism for the stent to be better seated within that calcified annulus. But we would exclude at present patients who have one big bunch of calcium at one commissure, for example. It needs to be equally distributed.
Dr Böning : May I make a comment? We started a pathological analysis in human aortic valves, and in 20–30% of the humans, the leaflets fall against the coronary ostia, if you push them towards the aortic wall. So you have a very high risk in a lot of patients to block the coronary arteries. I think we should start to work on the problem of valve leaflet removal.
Dr Walther : Well, it depends. There may be patients who have a somewhat enlarged sinus and the coronary ostia are a bit more away, the stent only dilates to 23 mm. There may be patients who require reoperative surgery as it was mentioned in your talk, and those who have open bypass grafts would maybe be ideal candidates for such an approach. It is just theoretical. We don’t have that much experience on that. It is just an experimental study so far.
Dr M. Deja (Katowice, Poland): Have you considered what happens if you implanted this in humans and then you need to remove it?
Dr Walther : Well, I think that is something you don’t want to have happen, but if it happens, of course, you have to go in and do the conventional approach, which is a very high risk, because I said our intention is to use such an approach for very high risk patients, even for patients whom some surgeons would consider as nonsurgical candidates. But of course, you then have to offer the patient some option because you don’t have a retrieval mechanism with this kind of prosthesis. Once it is deployed, it is deployed. That is a shortcoming, actually.
Dr R. Cesnjevar (Erlangen, Germany): You used rapid ventricular pacing for unloading and cardiopulmonary bypass. Which do you think is more effective? And secondly, as I understood the angiograms, the nitinol stent is not close to the sinuses or the wall. Do you expect any thrombus formation on this if it is just free-standing in the aortic root?
Dr Walther : Starting with the second, I don’t really expect thrombus because there is enough blood flow probably, and patients will get some aspirin. That should be fine. It is not a nitinol, it is a steel stent, so it doesn’t really close completely.
Regarding the unloading technique we realized that both are effective, when using rapid ventricular pacing the heart comes back, even in the animal; initially we were a bit anxious that they would keep on fibrillating but they came back, we used this technique because it is much easier. However, I personally think by now that both together is the best way to a hundred percent unload the heart so that you can position the valve at the very best place you want to have it.
Dr P. Kappetein (Rotterdam, The Netherlands): Do you think that a problem could occur with the chordae of the mitral valve, or haven’t you seen this?
Dr Walther : Well, with the mitral valve it really depends on the anatomy. In the animals we had about 2–3 mm distance below the annulus, the lower edge of the valve was sitting about 2–3 mm below the mitral valve annulus at the level of the anterior mitral leaflet. But the anatomy is very much different in the animals. We had a distance of the aortic annulus to the left coronary ostium of 7 mm, something like that, and in humans it is 12–15. So there is much more space, and in the other direction there should be more space as well. So I think it should be all right for human application.
Dr C. Mestres (Barcelona, Spain): You should define what is a real high risk patient, because EuroSCORE 9 is not really a great deal, at least in this part of the world.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Presented at the joint 19th Annual Meeting of the European Association for Cardio-thoracic Surgery and the 13th Annual Meeting of the European Society of Thoracic Surgeons, Barcelona, Spain, September 25–28, 2005. |
References |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. Discussion5. ConclusionAppendix AReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  H. A. Dwyer, P. B. Matthews, A. Azadani, L. Ge, T. S. Guy, and E. E. Tseng Migration forces of transcatheter aortic valves in patients with noncalcific aortic insufficiency J. Thorac. Cardiovasc. Surg., November 1, 2009; 138(5): 1227 - 1233. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Walther, V. Falk, M. A. Borger, J. Kempfert, J. Ender, A. Linke, G. Schuler, and F. W. Mohr Transapical aortic valve implantation in patients requiring redo surgery Eur. J. Cardiothorac. Surg., August 1, 2009; 36(2): 231 - 235. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. A. Dwyer, P. B. Matthews, A. Azadani, N. Jaussaud, L. Ge, T. S. Guy, and E. E. Tseng Computational fluid dynamics simulation of transcatheter aortic valve degeneration Interactive CardioVascular and Thoracic Surgery, August 1, 2009; 9(2): 301 - 308. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Abdel-Sayed, M. Kalejs, and L. K. von Segesser A new training set-up for trans-apical aortic valve replacement Interactive CardioVascular and Thoracic Surgery, June 1, 2009; 8(6): 599 - 601. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. E. McRae, M. Rodger, and B. A. Bailey Transcatheter and Transapical Aortic Valve Replacement Crit. Care Nurse, February 1, 2009; 29(1): 22 - 36. [Full Text] [PDF]
|
|
|
|
|
|
A. Zierer, G. Wimmer-Greinecker, S. Martens, A. Moritz, and M. Doss The transapical approach for aortic valve implantation. J. Thorac. Cardiovasc. Surg., October 1, 2008; 136(4): 948 - 953. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. K. Rosengart, T. Feldman, M. A. Borger, T. A. Vassiliades Jr, A. M. Gillinov, K. J. Hoercher, A. Vahanian, R. O. Bonow, and W. O'Neill Percutaneous and Minimally Invasive Valve Procedures: A Scientific Statement From the American Heart Association Council on Cardiovascular Surgery and Anesthesia, Council on Clinical Cardiology, Functional Genomics and Translational Biology Interdisciplinary Working Group, and Quality of Care and Outcomes Research Interdisciplinary Working Group Circulation, April 1, 2008; 117(13): 1750 - 1767. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. J. Davidson and D. S. Baim Percutaneous Aortic Valve Interventions Card. Surg. Adult, January 1, 2008; 3(2008): 963 - 971. [Full Text]
|
|
|
|
|
|
T. Walther, P. Simon, T. Dewey, G. Wimmer-Greinecker, V. Falk, M. T. Kasimir, M. Doss, M. A. Borger, G. Schuler, D. Glogar, et al. Transapical Minimally Invasive Aortic Valve Implantation: Multicenter Experience Circulation, September 11, 2007; 116(11_suppl): I-240 - I-245. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Walther, V. Falk, T. Dewey, J. Kempfert, F. Emrich, B. Pfannmuller, P. Broske, M. A. Borger, G. Schuler, M. Mack, et al. Valve-in-a-Valve Concept for Transcatheter Minimally Invasive Repeat Xenograft Implantation J. Am. Coll. Cardiol., July 3, 2007; 50(1): 56 - 60. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. J. Antunes Off-pump aortic valve replacement with catheter-mounted valved stents.: Is the future already here? Eur. J. Cardiothorac. Surg., January 1, 2007; 31(1): 1 - 3. [Full Text] [PDF]
|
|
|
|
|
|
T. Walther, V. Falk, M. A. Borger, T. Dewey, G. Wimmer-Greinecker, G. Schuler, M. Mack, and F. W. Mohr Minimally invasive transapical beating heart aortic valve implantation -- proof of concept Eur. J. Cardiothorac. Surg., January 1, 2007; 31(1): 9 - 15. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
