Transcatheter aortic valve implantation (TAVI) has recently emerged as a novel, minimally invasive treatment option for patients with symptomatic severe aortic valve stenosis deemed at high risk of conventional surgery due to excessive co-morbidities.1,2 Clinical results of TAVI are encouraging and the number of patients treated with this new method continues to grow exponentially in an unprecedented fashion, as does the number of new TAVI centres worldwide.3
Currently, two valve systems are approved for TAVI and commercially available in Europe. The Edwards Sapien (Edwards Lifesciences, Irvine, CA, US) valve is a trileaflet bovine xenovalve that is mounted in a balloon-expandable cobalt chromium stent (Sapien XT). The valve is deployed by balloon inflation under rapid right ventricular pacing at 160–220 beats per minute. The other available valve, the Medtronic/CoreValve Revalving system (Medtronic Inc., Minneapolis, MN, US) is a trileaflet porcine pericardial tissue valve mounted within a self-expandable nitinol frame that is deployed stepwise and under guidance by several small-volume angiograms without rapid pacing.
This article will review current developments in the field of TAVI, which are mainly directed towards further miniaturisation of catheter devices and sheaths for facilitated transfemoral access; development of retrievable and repositionable valve prostheses for facilitated implantation; and strategies for preventing embolic stroke during the procedure.
Miniaturisation of Valve Systems for Facilitated Transfemoral Access
Vascular access remains a major challenge in transfemoral (TF)-TAVI due to the need of large-bore introducer sheaths for valve delivery.4,5 The elderly patient population currently referred for TAVI often presents with advanced atherosclerosis of the iliofemoral axis. In a considerable number of patients with diffusely calcified access vessels with diameters below 7mm, TF-TAVI may not be possible at all and TAVI may be performed only transapically or via the subclavian/axillary artery. However, this increases procedural complexity and requires general anaesthesia. Even if TF-TAVI is possible, dissections, occlusions or ruptures of the iliofemoral arteries can occur in 13–32% of patients, significantly contributing to periprocedural morbidity and mortality.4,5
The issue of vascular access is usually considered to be less important for the Medtronic/CoreValve prosthesis, which requires ‘only’ an 18 French (Fr) arterial sheath. However, it has to be considered that the outer diameter (OD) of this 18Fr sheath is in fact 21Fr. Arterial access has been a major issue with the Edwards Sapien prosthesis, which required 22Fr (OD 25Fr) and 24Fr (OD 28Fr) sheaths for the 23 and 26mm valve, respectively, and, thus, a high number of patients had to be excluded from Edwards TF-TAVI for access-related reasons. Recently, Edwards has launched the new-generation Sapien XT valve prosthesis (see Figures 1 and 2). The cobalt chromium stent frame allows for a smaller crimping profile that is additionally reduced by crimping the valve not directly onto the delivery balloon, but on the catheter shaft. The valve is finally loaded onto the balloon after advancement to the descending thoracic aorta. This allows for a significantly reduced delivery profile requiring 18Fr for the 23mm and 19Fr for the 26mm valve. The sheath has a hydrophilic coating and can usually be advanced with less resistance, resulting in a felt additional diameter reduction of ~2Fr. Also, Medtronic is addressing the issue of vascular access and is currently working on the next-generation CoreValve prosthesis, which is promised to require only 16Fr introducer sheaths.
Recently, balloon-expandable sheaths have been developed and are currently commercialised to facilitate vessel access for TAVI. The Solopath® vascular sheath (Onset Medical Corporation, Irvine, CA, US) has a nominal inner diameter of 19Fr (OD 22Fr) and is compatible with the 18Fr Medtronic/CoreValve delivery system. The distal section of the sheath is folded over a length of 25cm with an outer diameter of approximately only 13Fr, allowing facilitated vessel entry (see Figure 3). After navigation of the sheath through the iliac axis to the aorta, a standard percutaneous transluminal coronary angioplasty (PTCA) balloon inflation device is connected to the sheath-integrated balloon and the folded section of the sheath is expanded to its nominal 19Fr diameter (see Figure 3B). Edwards is also developing an expandable sheath compatible with its Sapien XT/Novaflex device that can even be deflated after the procedure for easy withdrawal from the calcified vasculature.
Retrievable and Repositionable Valve Systems
Correct valve positioning is crucial for treatment success and optimal outcomes after TAVI. Implantation of the Medtronic/CoreValve prosthesis is performed stepwise under repeat angiographies. During the deployment, the catheter system has to be pulled back constantly to achieve an optimal position because the valve prosthesis is basically pushed out for about 5cm during withdrawal of the catheter housing. While the risk of ventricular embolisation is low with the Medtronic/CoreValve prosthesis, the valve can be ejected from the ventricle into the aorta during deployment, for example with frequent extrasystoles. As long as the prosthesis is only partially deployed, it can usually be withdrawn into the sheath, re-crimped and again implanted. Positioning of the Medtronic/CoreValve device in a position that is too low may result in significant aortic regurgitation due to incomplete seal and is a predictor of permanent pacemaker implantation after CoreValve-TAVI. The Edwards valve is implanted using balloon inflation under rapid pacing in a single-shot fashion. Therefore, positioning of the device as exactly as possible before balloon inflation is of utmost importance. Too low a position may result in ventricular embolisation of the valve; too high a position may result in coronary artery obstruction.
Deployment systems allowing safe valve retrieval and repositioning during any step of the procedure are receiving major interest in order to overcome the shortcomings of the current valve devices with respect to positioning. The Lotus® valve is a bovine pericardial tissue tri-leaflet prosthetic aortic valve that is sutured within a braided nitinol stent frame. The frame is surrounded by a flexible membrane to seal paravalvular gaps between the prosthetic and native valve. At the beginning of the procedure, the valve is constrained within a delivery catheter of 21Fr in a longitudinally stretched state with low radial force, which enables a small profile. After passage of the native valve, the Lotus valve is unsheathed to foreshorten and expand radially, thereby reducing its height and gaining radial force. The valve is already functional even if only partially implanted and therefore the absence of major aortic regurgitation or left ventricular obstruction during implantation allows for very controlled and precise deployment. Rapid right ventricular pacing is not required during valve implantation. Valve performance can be assessed in vivo before final release.
In case of unsatisfactory position/valve function, the valve can be re-sheathed and the position can be corrected, allowing both the advance and the retraction of the prosthesis for a deeper and more ventricular or shallower and more aortic position. The first implantation of the Lotus valve was performed in 2008.6 Currently, the second-generation Lotus valve is undergoing testing in a first-inhuman trial (see Figures 4 and 5).
The Direct Flow Medical aortic valve (Direct Flow Medical, Inc., Santa Rosa, CA, US) is the first transfemorally delivered aortic valve prosthesis that is not based on a metallic stent frame technology. It allows for retrieval, repositioning or sizing exchange before permanent implantation. First results have recently been published.7 However, severe leaflet and left ventricular outflow tract calcification represents a major limitation for this prosthesis and, as such, implantation of the Direct Flow valve was only possible in 22 of 31 attempted patients.7
Prevention of Periprocedural Embolic Stroke
TF-TAVI bears the risk of periprocedural stroke with stroke rates reported as high as 2.9–10%.8 Advancement of semi-rigid, large-bore delivery catheters through the aortic arch, as well as direct manipulation of the usually heavily calcified aortic valve during passage and prior balloon valvuloplasty and, finally, crushing of the stenosed native leaflets by implantation of the metallic stent frame containing the xeno-pericardial tissue valve into the aortic annulus, may result in dislodgement of micro-debris from arch atheroma or from the valve itself with a risk of subsequent embolic stroke.8 Previous work from our group using diffusion-weighed magnetic resonance imaging (MRI) has underlined this issue, demonstrating multiple new embolic cerebral lesions in almost all patients after TAVI, although these foci were not associated with apparent neurological events or measurable deterioration of neurocognitive function during three-month follow-up.8 Nevertheless, efforts have been directed towards prevention of embolic stroke and cerebral protection devices have been developed particularly for use during TAVI.
The Embrella Embolic Deflector® (Embrella Cardiovascular Inc., Wayne, PA, US) is such an embolic protection device specifically designed to reduce the amount of embolic material that may enter the carotid arteries during TAVI. The Embrella Embolic Deflector is placed in the aorta via a 6Fr sheath through a right brachial or right radial artery access site (see Figure 6).
The distal end of the Deflector is comprised of an oval-shaped nitinol frame covered with a porous membrane (100μm). The frame has two opposing petals that are positioned along the greater curvature of the aorta, covering the ostia of both the brachiocephalic and the left common carotid arteries. The porous membrane permits blood flow to the carotid arteries while simultaneously deflecting embolic material away from these vessels during the index procedure. The Embrella Embolic Deflector has been tested in a first-in-man multicentre trial and the results are currently awaited.
Conclusion
TAVI has become a viable option for treatment of patients with severe symptomatic aortic valve stenosis. Current developments are promising to facilitate the TAVI procedure technically and improve procedural safety, thereby improving patient outcomes. Miniaturisation of catheter devices and sheaths will allow the treatment of more patients via the transfemoral route. Retrievable and repositionable valve systems will improve acute functional and haemodynamic valve results and presumably improve valve durability, which will be of great importance once the indication for TAVI is broadened to include younger patients and fewer high-risk patients. In all patients, prevention of periprocedural embolic stroke is of utmost importance and current developments of embolic deflectors are promising to improve patient safety.