Transcatheter mitral valve replacement in degenerated calcified native mitral valves: is the currently available technology suitable?

In the last decade, transcatheter heart valve (THV) procedures in adults have been gaining more and more popularity in Western countries, and nowadays, inoperable patients or patients at high risk for standard open-heart surgery can undergo transcatheter aortic valve replacement (TAVR) with last-generation balloon-expandable or self-expandable transcatheter aortic bioprosthesis. Multiple clinical trials have reported favourable outcomes with low morbidity and mortality rates. For this reason, TAVR has been accepted by many professional societies as an alternative to surgery in high-risk patients [ 1 , 2 ]. Currently, THV implantations are possible in stenotic calcified native aortic valves, in selected cases of regurgitant native aortic valves using the JenaValve™ (JenaValve Technology, Inc., Irvine, CA), stenotic and calcified pulmonary conduits, and in degenerated stented or stentless bioprosthesis regardless of the anatomic position [ 1–6 ]. As for the atrio-ventricular valves, the native mitral valve represents a challenge for transcatheter therapies because of the more complex anatomy and the subjacent disease involving the leaflets and the subvalvular apparatus. However, transcatheter devices for the treatment of mitral valve disease, such as the MitraClip™ (Abbott Vascular, Menlo Park, CA) and the NeoChord™ system (NeoChord Inc., Eden Praire, MN), have been already in use for years, and newer devices, such as the Valtech Cardioband™ (Valtech Cardio, Or, Yehuda, Israel) percutaneous annular ring, have recently obtained the conformity marking from the European Community (CE-mark approval) for clinical use in Europe [ 7 , 8 ]. Promising devices for transcatheter mitral valve replacement (TMVR) in mitral regurgitation have already started (or are ready to start) early feasibility clinical trials: CardiAQ™ (CardiAQ Valve Technologies, Irvine, CA), Tendyne™ (Tendyne Holdings, Roseville, MN), Twelve™ (Twelve, Inc., Redwood City, CA), Tiara™ (Neovasc Inc., Richmond, BC) and Fortis™ (Edwards Lifesciences, Irvine, CA). However, they all are in very early stages of evaluation and will not be commercially available in the very near future. In addition, their role in severe mitral annular calcification (MAC) may be limited as these self-expandable devices are specifically designed for non-calcified pathology and may not have the radial strength needed to treat massively calcified stenotic mitral valves [ 9–11 ]. In frail elderly patients, calcified stenotic mitral valves can be surgically addressed during standard on-pump open-heart procedures, but the technique carries a well-known risk of calcium embolization and atrio-ventricular rupture/disjunction during the extensive decalcification, and the surgical risk is even higher in patients with comorbidities. Therefore, non-surgical (or extremely high-risk) patients have been treated with the compassionate use of aortic THV devices and transapical/transfemoral delivery systems for aortic valve replacement (Edwards Sapien™, Lotus™ Boston Scientific, Direct Flow™ Medical). Several isolated successful implants have been published [ 12–18 ]. Since 2013, an international Global Registry for TMVR in MAC has been collecting outcomes from patients treated at multiple centres in North America, Europe and South America [ 19 ]. TMVR in MAC has been performed through different access sites and using different surgical or cardiac set-ups in the cath lab, hybrid suite or operation theatre (Fig. 1 ). The transapical access allows for a straightforward beating-heart mitral valve replacement and, in the Global Registry, this has been the most frequently used access site (45%). Using this approach, balloon-expandable THVs can be implanted in the mitral valve using the transapical system with the valve mounted upside down compared with standard transapical aortic valve replacement (Fig. 1 B). This technique facilitates a more coaxial valve alignment with stability of the delivery system during deployment compared with a less coaxial and less stable position often seen in transseptal delivery. However, operators must have caution to prevent vascular injury when stiff guidewires are placed in the left atrium and fragile pulmonary veins [ 12 , 13 ]. Self-expandable aortic THVs designed for transfemoral delivery (Boston Lotus™ valve and Direct Flow™ valve) have been successfully implanted in calcified native mitral valves utilizing the transapical access route [ 14 , 15 ]. The main advantage of these valves is that they can be retrieved and repositioned before they are released [ 14 , 15 ]. These features are particularly useful when significant left ventricular outflow tract (LVOT) obstruction is seen after valve deployment. ( A ) An open-heart transatrial transcatheter mitral valve replacement (TMVR) with a 29-mm Sapien™ 3 placed in the heavily calcified mitral valve after anterior leaflet removal. ( B ) The post-TMVR echocardiographic control showing good valve functioning without LVOT obstruction. ( C ) A transfemoral–transseptal TMVR with a Sapien™ 3 valve. The transvenous transseptal access site is also a frequently used route (41% of the cases in the early experience of the global registry). The balloon-expandable aortic THVs have been used with this approach and mounted upside down on the delivery system (Fig. 1 C and D) [ 16 , 17 ]. In some cases, a wire has been externalized through a sheath percutaneously placed in the left ventricle to provide more coaxiality and support during valve deployment [ 16 ]. Direct implants of Sapien™ stent-valves through an open transatrial approach via a right thoracotomy or a median sternotomy (on cardiopulmonary bypass with cardioplegic cardiac arrest or fibrillating heart) have been performed and reported with good clinical and haemodynamic results (14% in the Registry) [ 18 , 19 ]. During the procedure, the stent-valve is mounted in the standard fashion used for transapical aortic valve replacement and the implant is performed under direct vision. The prevention of valve migration is obtained by fixing the valve to the annulus with three stitches (Fig. 1 A) [ 18 ]. A great advantage of this approach is that the anterior mitral leaflet (or part of it) can be removed to reduce the risk of LVOT obstruction after stent-valve deployment. Another access route option for TMVR is the left transatrial delivery through a right mini-thoracotomy: this way is similar to the surgical one and can be performed with transapical Sapien™ valves and delivery systems, but the Certitude™ sheath is placed through the wall of the left atrium (with pledgeted purse-string sutures between the right pulmonary veins) after the pericardial opening and without cardiopulmonary bypass use. Balloon-expandable Edwards Sapien™ valves have been the most frequently used devices for transcatheter mitral implantation in MAC. The last-generation Sapien™ 3 valve has a cobalt-chromium stent frame with sewn bovine pericardial leaflets available in four sizes (20, 23, 26 and 29 mm) and featuring two new characteristics: (i) large open cells on the outflow that facilitates access to coronary arteries when needed and (ii) an innovative skirt system on the inflow designed to reduce paravalvular leak. Although the skirt was designed to reduce paravalvular aortic regurgitation, it may help reduce paravalvular mitral regurgitation when implanted in the mitral position. An important advantage of the Sapien™ is that the valve can be mounted on the balloon-catheter upside down, allowing the operators to utilize a variety of delivery approaches (transatrial, transapical and transseptal). The Lotus™ valve (Boston Scientific, Malborough, MA, USA) is a self-expandable nitinol valve available in three sizes (23, 25 and 27 mm) with the advantage of being fully retrievable and repositionable before release. However, the valve is already pre-mounted in one direction for transfemoral aortic delivery. Therefore, the transapical access is the only delivery method for implantation in the mitral position at this time. Lim et al . [ 15 ] reported the first two successful implants in native valves using this approach. The advantage of this system is that the valve can be repositioned or retrieved in case of LVOT obstruction or unacceptable paravalvular leakage. The Direct Flow™ (Direct Flow Medical, Santa Rosa, CA, USA) is a self-expandable, repositionable and metal-free frame made in polyester with bovine pericardial leaflets, available in four sizes (23, 25, 27 and 29 mm). Similar to the Lotus valve, the Direct Flow valve is designed for transfemoral delivery in the aortic position requiring transapical access for delivery in the mitral valve. Repositionability and retrievability characteristics are advantageous when LVOT obstruction or significant paravalvular leak is seen after TMVR. Operators have extrapolated methods used for aortic annulus sizing to size the mitral valve annulus. Cardiac computed tomography (CT) has been the most reliable method. The minimum and maximum diameters, and mitral annular area and perimeter are measured. Similar to TAVR with balloon-expandable valves, the area is the most frequently used measure to determine the size of balloon-expandable THV used in the mitral position. Accurate sizing of the mitral annulus is more difficult than for the aortic valve due to the complex morphology with a non-planar, saddle-oval shape. The most difficult segment to measure is the anterior aspect or aorto-mitral continuity because there are important inter-reader and intra-reader variabilities when measuring this segment, due to the dynamic non-planar shape. Blanke et al . [ 20 ] suggested a method that eliminates this segment when calculating the mitral annular area, converting the saddle shape into a D-shape that starts at the medial and lateral trigones. Although the D-shape concept may be less relevant when the aorto-mitral continuity is calcified as it is in severe MAC cases, the method can be used in this patient population as well (Fig. 2 ). ( A ) CT-based measurements of the heavily calcified mitral annulus. ( B ) Hockey-puck maximum intensity projection image of calcium using 3Mensio Structural Heart Mitral Workflow software (Pie Medical Imaging, Maastricht, Netherlands). Accurate measurement is extremely important for pre-procedural planning. Oversizing may increase the risk of LVOT obstruction and, in theory, the risk of annular rupture as well, although the latter has not been reported yet. On the other hand, undersizing may result in valve embolization. This concept is particularly important in the mitral position where there is a much larger difference in pressures between chambers when the valve is closed compared with the aortic valve. The embolization rate reported in the initial analysis of the global registry was 6% (4 out of 64) [ 19 ]. In all these cases, the valve embolized into the left atrium. The embolizations occurred in the very early experience and the embolization rate has improved with better sizing methods. In direct transatrial TMVR cases, fixing the valve to the annulus helps prevent valve migration or embolization but sizing remains important to minimize paravalvular regurgitation. Post-procedural LVOT obstruction is a potential life-threatening complication after TMVR in MAC. The rigid calcified anterior mitral leaflet is pushed towards the LVOT by the transcatheter valve, and therefore, there is a risk of increased sub-aortic gradient in patients with narrow LVOT or asymmetric hypertrophic interventricular septum. The incidence rate of severe LVOT obstruction with haemodynamic compromise was 9.3% (6 out of 64) in the initial report of the Global registry [ 19 ]. This serious complication was associated with adverse outcomes and in-hospital mortality in 5 of these patients despite aggressive measures taken to improve the LVOT obstruction. Careful CT analysis helps identify risk factors for LVOT obstruction: left ventricular hypertrophy particularly with symmetric increased septal thickness, unfavourable aorto-mitral angle, small ventricular cavity and long anterior mitral leaflet (Fig. 3 ). Other important factors are the selected valve size and the depth of implantation in relation to the mitral annulus, for which CT is also helpful. The larger is the THV size and the more ventricular is the implantation, the higher is the risk of LVOT obstruction. ( A ) CT-scan image showing asymmetric septal hypertrophy, a long anterior mitral valve leaflet and a small left ventricular outflow tract (LVOT) in systole, suggesting a high risk of obstruction. ( B ) Measurement of the neo-LVOT area with a virtual valve in place simulating the new LVOT area post-TMVR utilizing 3Mensio Structural Heart Mitral Workflow software (Pie Medical Imaging). TMVR: Transcatheter mitral valve replacement. In direct transatrial TMVR, the anterior mitral leaflet can be removed: this is a useful alternative for patients at risk for LVOT obstruction during transseptal or transapical TMVR. Although this is a more invasive approach requiring a small right thoracotomy at the fourth intercostal space, peripheral cardiopulmonary bypass cannulation and fibrillating heart, this may have a lower associated risk than standard surgery in the presence of a massively heavily calcified mitral annulus [ 18 ]. Although the initial case reports were encouraging regarding the technical feasibility of TMVR, the safety and efficacy of this procedure as well as the incidence of complications were unknown. The TMVR in MAC Global Registry was created to collect outcome data of similar procedures performed in centres with experience of TMVR in MAC, and the outcomes of the first 64 patients who underwent TMVR with the compassionate use of a balloon-expandable THV were presented at Transcatheter Cardiovascular Therapeutics meeting in 2015 [ 19 ]. The mean age was 73 ± 13 years (66% female), mean STS score was 14 ± 9.5 and mean left ventricular ejection fraction was 59 ± 11%. The valve was stenotic in 93.5% and regurgitant in 6.5% (MR) of patients. The mean mitral valve gradient (MVG) in patients with stenosis was 11 ± 4.4 mmHg with a mean mitral valve area (MVA) of 1.18 ± 0.51 cm 2 . Most patients were in NYHA class III or IV (92%). Transatrial delivery under direct visualization through an open surgical approach was utilized in 15.6% of the cases, whereas the transapical and transfemoral–transseptal approaches were used in 43.8 and 40.6% of the cases, respectively. The rate of procedural success was 72%, primarily limited by the need for a second valve in 11 patients (for migration in 5 and regurgitation in 6). Post-procedural MVG was 4 ± 2.2 mmHg, mean MVA was 2.2 ± 0.95 cm 2 and paravalvular regurgitation was mild or absent in all. Four valves embolized in the left atrium (6.25%) and 6 patients (9.3%) had severe LVOT obstruction with haemodynamic compromise after valve deployment. The hospital mortality rate was 29.7% for cardiovascular (12.5%) and non-cardiac (17.2%) causes. Cardiovascular deaths were due to LVOT obstruction (2), left ventricular perforation (2), stroke (2), complete AV block (1) and acute myocardial infarction secondary to air embolism following guidewire-induced pulmonary vein perforation (1); non-cardiac deaths were due to multi-organ failure (5), pneumonia (3), infection (2) and bleeding (1). Thirty-day follow-up echocardiographic data were available in 22 patients with a mean MVG of 5.9 ± 2.1 mmHg and mean MVA of 2.3 ± 0.8 cm 2 . Eighteen patients (81.8%) had zero/trace MR and 4 (18.2%) had mild MR; moderate/severe MR was absent in all. Survivors reported improvement in symptoms: 21 of the 25 patients (84%) with 30-day clinical follow-up were in NYHA class I or II, and 4 (16%) were in NYHA class III. TMVR might evolve into an acceptable alternative for selected patients with severe mitral calcifications who are not candidates for conventional heart surgery. However, this field is at a very early stage and important challenges exist with the currently available technology developed to treat aortic valve diseases. In particular, more data are needed to determine better methods for valve sizing, prevention of embolization and for the prevention of LVOT obstruction: for that reason, the MITRAL trial (Mitral Implantation of TRAnscatheter vaLves), a physician-sponsored FDA-approved IDE trial aiming to evaluate the safety and feasibility of Sapien™ valves in inoperable patients with MAC, was recently initiated (ClinicalTrials.gov Identifier NCT02370511). This trial will provide insights to improve technical success, patient selection process and the overall clinical outcomes of this very interesting but still controversial field. Conflict of interest: Enrico Ferrari and Danny Dvir are consultants for Edwards Lifesciences. Mayra Guerrero is a proctor and receives research grant support from Edwards Lifesciences.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.