Transcatheter Mitral Valve Replacement: Evolution and Future Development

We will review transcatheter mitral valve replacement (TMVR) and discuss this evolving cutting edge procedure in terms of types (valve in valve, valve in ring and valve in mitral annular calcification MAC), clinical indications, pre-procedural planning and value of pre-procedural imaging including computed tomography role, technical challenges encountered in these procedures, potential complications for each type of TMVR, and potential strategies to mitigate and avoid such complications, We will review the currently available devices dedicated for mitral valve replacement, with a summary of their preliminary data and early outcome results. We will also discuss knowledge gaps and ideas for future research.


Introduction
Valvular heart disease affects >100 million patients worldwide, which is estimated to increase further with the aging population and a subsequent increase in degenerative valve disease [1]. Based on analysis of the Society of Thoracic Surgery (STS) National Database, there are >40,000 mitral valve replacements being performed annually in the United States (US), with shift from mechanical to bioprosthetic valve replacements [1]. It is known that redo mitral valve surgery is associated with higher mortality compared to first mitral surgery; with 30-day mortality ranging from 6% for elective second mitral valve surgery and 17.8% for emergency surgery [2]. The risk of a third or fourth surgery is even higher; with 30-day mortality reaching up to 44% in urgent surgery [3]. As such, Transcatheter mitral valve replacement (TMVR) using aortic balloon-expandable transcatheter heart valves (THV) has been increasingly performed for patients with severe mitral valve disease who are not candidates for surgery [4]; as it emerged as a less invasive alternative option for these patients with relatively lower mortality than the predicted STS predicted rates of mortality [4]. Moreover, dedicated devices for TMVR have been developed and some are currently being studied [5][6][7][8][9][10]. Results of the clinical outcomes of TMVR are promising, but anatomical differences between mitral bioprosthetic valves, annuloplasty rings, and severely calcified mitral annulus are associated with specific procedural challenges for TMVR procedures [1,4].
The role of mitral ViV, ViR, and ViMAC has been evaluated in a prospective early feasibility clinical trial, the MITRAL trial (Mitral Implantation of Transcatheter Valves, NCT 02370511), which is the first prospective study assessing outcomes of TMVR in all of the three separate subtypes. The results of the trial have been recently published [16][17][18].

Author, year published Number of patients Major outcomes
Guerrero et al. [11] 64 patients with ViMAC • Technical success was 72% • 30-day all-cause mortality was 29.7% • 84% of the survivors with follow-up data available were in New York Heart Association (NYHA) class I or II at 30 days Yoon et al. [13] 248 In the MITRAL trial, in which 30 patients undergoing transseptal mitral ViV were enrolled between July 2016 and October 2017, technical success was achieved in 100% of cases with 30-day all-cause mortality of 3.3%, which remained unchanged at 1 year. At 1-year follow-up, the vast majority of patients were in New York Heart Association (NYHA) functional class I or II [16].
Similarly, in the MITRAL trial assessing patients undergoing transeptal mitral ViR, 30 patients were studied with results showing technical success of 66.7% (driven primarily by need for a second valve in 6 patients), all-cause mortality of 6.7% at 30 days and 23.3% at 1 year. Similar to ViV study, the vast majority of patients were in NYHA class I or II at 1 year [17].

Procedural planning
Successful TMVR depends on accurate sizing of the mitral annulus and avoidance of LVOT obstruction. In the absence of a validated standard method for mitral annulus sizing at the present time, operators have extrapolated from transcatheter aortic valve replacement (TAVR) experience and used a variety of sizing approaches including echocardiography, 3-dimensional (3D) transesophageal echocardiography, cardiac CT, and balloon sizing techniques [10]. Cardiac CT is the most accepted imaging modality for annulus sizing. In general, pre-procedural imaging constitutes of contrast-enhanced CT to identify critical cardiac structures and anatomy, including sizing of the mitral annulus, which is the basal-most structure of the mitral leaflets [19]. In addition to annular sizing, CT also provides essential information for pre-procedural planning, including the amount and distribution of calcifications, as well as predictors of LVOT obstruction; the left ventricular cavity size, anterior leaflet length, aorto-mitral angulation, septal hypertrophy, among other features. CT is also helpful in identifying the trajectory and site of access, whether transapical or transseptal [10,19].
Data utilizing 2-dimensional (2D) echo imaging correlated acute angulation of the mitral aorta-outflow-angle (mAOA) with higher risk of LVOT obstruction compared with that of more obtuse mAOA. However, risk of LVOT obstruction is not solely based on mAOA; this is because LVOT is a 3D anatomical structure and mAOA on 2D echo images may not provide the comprehensive assessment needed. CT overcomes this limitation as it provides a 3D assessment. Both the prosthetic valve and the anterior displacement of anterior mitral leaflet can result in severe LVOT obstruction. Additionally, utilization of computer-aided designs and 3-D printed models allows us to test devices in patient-specific anatomy and at different angulations and depths with estimation of risk for LVOT obstruction [19].
LVOT obstruction is a fatal complication; thus, pre-procedural planning in an attempt to predict neo-LVOT provides a key step in the success of TMVR procedure. In a multicenter study of 38 patients undergoing TMVR using balloon-expandable valves for severe mitral valve dysfunction because of degenerative surgical mitral ring, bio-prosthesis, or severe native mitral stenosis from severe mitral annular calcification, the investigators defined LVOT obstruction as increase of 10 mmHg or more in LVOT peak gradient following TMVR and found that 7 of the 38 patients had LVOT obstruction, with CT neo-LVOT surface area correlating well with measurements after TMVR [20]. Yoon and colleagues in their study of 194 patients undergoing TMVR found that LVOT obstruction was associated with higher procedural mortality compared with patients without LVOT obstruction (34.6% vs. 2.4%; p < 0.001) [21].
Because the mitral annulus is larger in size compared to aortic valve annulus, TMVR requires larger devices, including prosthesis and delivery systems [10]. Mitral annular calcifications are less common compared with aortic valve calcifications, and their presence may condition the implant of a transcatheter mitral prosthesis. For this purpose, the role of TMVR in presence of considerable annular calcification is less clear, as shown in the MAC (mitral annular calcification) Global Registry, which demonstrated that TMVR was feasible in MAC but associated with relatively high early and midterm mortality at 1 year, although patients who survived at 1-year follow-up had sustained improvement of symptoms [4,5]. Similarly, the MITRAL trial showed relatively high 1-year mortality in ViMAC patients, but transeptal ViMAC showed promising results with 30-day mortality lower than the predicted STS score, however mortality rates in this population remains higher than other transeptal procedures, including transcatheter edge-to-edge repair using MitraClip [1,2,18].

ViV
Complications in ViV are considered relatively low, with reported LV perforation 0.4%, LVOT obstruction 0.7% and conversion to surgery in 1.3% [1,4]. Post-procedure mitral valve function was excellent with a median mean mitral valve gradient of 4 mm Hg and residual mitral regurgitation grade of 1+ or less in 98.1%. A second valve was needed in a relatively small proportion of mitral ViV patients (1.5%) and was associated with higher mortality at 30 days. The reasons or mechanisms for which this was associated with higher mortality (residual mitral regurgitation, thrombosis, renal failure) are not known at this time [4].

VIR
Generally speaking, studies have shown that ViR TMVR is associated with worse outcomes compared with ViV, but better outcomes compared with ViMAC procedures [1,4,10,[16][17][18]; ViR is a more complex procedure than ViV due to the different types of rings (rigid versus nonrigid, complete versus incomplete) and different shapes, which are usually not round predisposing to residual paravalvular leak [1,4]. There are 3 main challenges in ViR cases: valve anchoring, LVOTO and paravalvular leak. Yoon et al. showed that ViR had a significantly lower technical success rate compared with the ViV group (83.3% vs. 96.0%; p = 0.001) due to more frequent second valve implantation (11.1% vs. 2.8%; p = 0.008) [1]. Moreover, the investigators found that residual mitral regurgitation moderate or higher at 30 days was more frequent in patients with flexible rings compared with those with semi-rigid rings (44.4% vs. 10.8%, p = 0.02) [1]. A study showed that the 30-day mortality was 11.5% in ViR patients with median STS PROM score of 9.3% [4]. The reasons for higher mortality in mitral ViR are probably multifactorial; potentially related to higher procedural complication rates including LVOT obstruction, higher valve embolization rate, residual mitral regurgitation and need for reintervention including conversion to surgery, as well as different baseline characteristics including a lower baseline left ventricular ejection fraction [4]. In fact, the ViR group also had the highest rate of device embolization at 30 days 3.6% compared with mitral ViV 0.2% and ViMAC 1.6% [4].
Guerrero et al. showed 4.9% rate of LVOT obstruction in ViR, which was lower than the 8% in the VIVID registry; this could be related to increased experience in patient selection and risk-reduction strategies. Overall, mitral ViR is observed to have higher rates of LVOTO as compared to ViV, possibly due to the presence of a preserved anterior mitral leaflet. In most ViV cases, the anterior leaflet is no longer present making LVOT obstruction less likely [4].
Guerrero et al. also demonstrated that when comparing outcomes by types of rings (complete versus incomplete, rigid versus nonrigid), there was a larger mitral valve area in incomplete rings versus complete rings [4]. However, there was no statistically significant difference in median mean mitral valve gradients and clinical outcomes between the groups based on the type of ring [4].

ViMAC
Studies have shown that ViMAC procedures were associated with the lowest technical success and the highest in-hospital and 30-mortality compared with mitral ViR and ViV [1,4]. Similar to ViR, ViMAC has significant challenges to anchoring, paravalvular leak and LVOTO. The reasons are multifactorial, including presence of multiple comorbidities and technical challenges including the complexity of the mitral valve anatomy; as the native mitral valve is a saddle oval shape being treated with a round transcatheter valve which may lead to paravalvular leak at commissures, non-uniform calcium distribution, and relatively small sized ventricles accompanied by threatened LVOTO [1,4,10]. Therefore, there is a frequent need for LVOT modification taking the form of three options: LAMPOON, Alcohol septal ablations or surgical resection of the anterior leaflet.
LVOT obstruction is considered the Achilles' heel of TMVR, especially in ViMAC. It has limited treatment options and was the strongest predictor of 30-day and 1-year mortality in the TMVR in MAC Global Registry [4,5,10]. Studies have shown that LVOT obstruction rate in ViMAC procedures is at least 10% [1,4,5,10]. One factor that could contribute to different rates of LVOT obstruction observed among registries may be the different definitions used, such as LVOT obstruction with hemodynamic compromise versus increase in mean LVOT gradient of ≥10 mm Hg from baseline. Another important factor may be improved screening process with cardiac computed tomography to predict LVOT obstruction and strategies to prevent it [1,4]. Potential predictors of LVOT obstruction are the angle of the mitral valve in relation to the LVOT long axis, the presence of small LV cavity, bulging or severe hypertrophy of the basal interventricular septum, long anterior mitral valve leaflets, dynamic alterations as the pushing of the native anterior leaflet toward the LVOT, prosthesis protrusion and device flaring [4,5,10,[19][20][21].

Strategies to mitigate procedural complications
Cardiac computed tomography to measure the expected neo-LVOT area to assess the risk of TMVR-induced LVOT obstruction identifying patients at risk facilitates implementation of measures to decrease such risk including preemptive alcohol septal ablation, percutaneous laceration of the anterior mitral leaflet, surgical excision of the anterior mitral leaflet during trans-atrial TMVR or deciding not to perform the procedure at all [1,4,10,24,25].
Several strategies to prevent or treat LVOT obstruction caused by TMVR have been developed and studied. These strategies include: 1) preemptive alcohol septal ablation in patients at risk for TMVR-induced LVOT obstruction who have favorable anatomy for alcohol ablation as shown in the MITRAL trial [18], 2) percutaneous DOI: http://dx.doi.org/10.5772/intechopen.98953 laceration of the anterior leaflet to decrease the risk of TMVR-induced LVOT obstruction in TMVR procedures (The LAMPOON trial; Laceration of the Anterior Mitral Leaflet to Prevent Outflow Obstruction During TMVR) [24,25], 3) possibly trans-atrial surgical access for TMVR in severe MAC, as evaluated by the SITRAL Trial (Surgical Implantation of Transcatheter Valves) [1,4,24,25].

Dedicated TMVR devices
In addition to the balloon expandable valves which were initially designed for the aortic valve and have been used in mitral valve interventions, several valve designs dedicated to the mitral valve have been developed and studied in several studies with a relatively small number of patients, with some promising results [10]. These dedicated mitral valves are summarized in Figure 1 and Table 2.
The CardiAQ (Edwards Lifesciences Inc) valve is a nitinol self-expanding trileaflet valve, composed of bovine pericardial tissue, which was the first dedicated device for TMVR in 2012 in high-risk patients with severe MR. This was followed by the second generation of the valve, which was used for the first time in 2014 [26]. The new redesigned version was renamed as the EVOQUE valve. It offers both a transapical and transfemoral-transseptal approach. The EVOQUE valve offered enhanced maneuverability and depth control, and lower ventricular projection to avoid LVOT obstruction. Currently, the Edwards EVOQUE TMVR Early Feasibility Study (NCT02718001) is recruiting and will assess feasibility at 30 days. The RELIEF (Reduction or Elimination of Mitral Regurgitation in Degenerative or Functional Mitral Regurgitation With the CardiAQ-Edwards™ Transcatheter Mitral Valve, NCT02722551) trial was stopped by the Edwards Company for further design validation. From the preliminary results presented, 13 patients have been treated with technical success of 92% and high mortality rate of 45% at 30 days [10,27].
The Tiara (Neovasc Inc., Canada) valve is a bioprosthetic valve; it constitutes of bovine pericardial tissue, which is mounted inside a nitinol frame. It is selfexpanding and has a relatively large atrial skirt, which decreases the risk of paravalvular leaks. The first implant of Tiara valve was performed in Vancouver in 2014. The two major studies of the Tiara valves, TIARA-I (Early Feasibility Study of the  [10]. Neovasc Tiara Mitral Valve System) (NCT02276547) and the latest TIARA-II (Tiara Transcatheter Mitral Valve Replacement Study), are actively enrolling patients, with promising preliminary results in 71 patients, mostly in functional MR (61%), showing 94% technical success rate, with a mortality rate of 11.3% at 30 days [28,29]. The FORTIS (Edwards Lifesciences Inc) valve is a self-expanding bioprosthetic valve of bovine pericardial tissue. The first FORTIS implant was performed in 2014. Preliminary results demonstrated outcomes of 13 patients with procedural success of 76.9%. In the early experience, the study was put on hold due to reported valve thrombosis [30]. The Tendyne MV system (Abbott Inc) is a self-expanding porcine pericardial valve, which is mounted on a nitinol stent. It is implanted using the transapical approach and the device is anchored to the annulus using apical tethers. The first Tendyne MV implant was performed in 2014. The Feasibility Study of the Tendyne Mitral Valve System for Use in Subjects With Mitral Annular Calcification (NCT03539458) of the first 100 patients showed that the technical success rate was 97% with no periprocedural mortality and 30-day mortality rate of 6% [31]. Most patients (98.8%) had non-significant MR at 30 days [31]. Importantly, the SUMMIT (Clinical Trial to Evaluate the Safety and Effectiveness of Using the Tendyne Mitral Valve System for the Treatment of Symptomatic Mitral Regurgitation; NCT03433274) is an ongoing multi-center clinical trial randomizing patients to TMVR using the Tendyne valve versus conventional mitral valve surgery, with goal to enroll 1010 patients and expected completion year in 2026.
The Intrepid (Medtronic Inc) valve is a bovine pericardial valve fixed onto a self-expanding nitinol frame. The valve is implanted via transapical access in the majority of cases, but a transseptal approach is being developed. The first implant was performed in 2014. An initial pilot study enrolled 50 high-risk patients with MR and reported 96% success rate, 14% 30-day mortality rate and trivial-trace or mild residual MR at 30 days in all patients [32]. The multicenter randomized APOLLO (Transcatheter Mitral Valve Replacement With the Medtronic Intrepid TMVR System in Patients With Severe Symptomatic Mitral Regurgitation; NCT03242642) trial is still ongoing with patients randomized in a 1:1 fashion to the Intrepid valve versus conventional mitral valve surgery. Outcomes will be evaluated at 30 days, 6 months, and 1 year, with up to 5 years follow-up duration, and estimated study completion date in 2025 [10].
The HighLife (HighLife SAS, France) valve is a 2-component system. The prosthetic valve is implanted in the mitral position and has an anchoring system placed by the trans-arterial retrograde approach in the sub-annular position [10]. Anecdotal initial cases are reported showing acceptable results of the valve [33]. A feasibility trial (NCT02974881) is still active.
The SAPIEN M3 (Edwards Lifesciences Inc) valve is a modified SAPIEN 3 valve. It is implanted using the transseptal approach. It uses nitinol docking system, which allows anchoring of device [10]. From the initial results of feasibility study in 15 patients, technical success was achieved in 86.7%, MR reduction was achieved in  93.3% and no mortality was reported at 30 days [34]. The ENCIRCLE trial is an ongoing study designed to assess the outcomes of the SAPIEN M3 device in 400 patients (NCT04153292). The Cardiovalve (Cardiovalve, Israel) valve is a bovine pericardial valve mounted on dual nitinol. It is usually implanted using the transseptal approach. The system allows multi-steerable catheter utilization, which provides better control of the device [10]. The AHEAD (European Feasibility Study of the Cardiovalve Transfemoral Mitral Valve System; NCT03339115) study is designed to assess the outcomes of Cardiovalve system in MR. The first 5 cases showed 100% of technical success with significant reduction of MR and absent or non-significant paravalvular leak [35].
The Cephea (Cephea Valve Technologies) system is a repositionable and recapturable frame valve and usually implanted using the transseptal approach. The frame structure allows adequate anchoring independent of the sub-valvular structures. The valve was tested in preclinical models with good performance at 90 days [36]. In addition, early experience with the Cephea device has been reported in 3 patients after the first in-human case with 100% technical success [37,38]. After a median 6-month follow-up, valve function, echocardiographic parameters and patients' functional status were all favorable [38].
The AltaValve (4C Medical Technologies Inc) is a supra-annular device, which constitutes of bovine tissue mounted onto a nitinol frame. It is a self-expanding valve and implanted using the transseptal or transapical approach. Animal studies showed good performance, and a first-in-human case was performed in Canada in 2018, with satisfactory results [10,39].
The NaviGate (NaviGate Cardiac Structures Inc) valve is a self-expanding valve and constitutes of a nitinol stent frame and multiple annular winglets, to allow anchoring of the device in the mitral annular position. The valve is implanted using transapical approach. The first in-human valve implant was performed in 2015 in Chile [40]. After an initial interest of this valve in mitral valve position, the device was implanted in the tricuspid position for tricuspid regurgitation using transcatheter interventions with trans-jugular or trans-atrial approach [41].

Knowledge gaps and ideas for future research
With the recent advances in TMVR in the most recent years, there remain knowledge gaps and challenges in order to understand the disease and correlate clinical outcomes with this evolving technology. MR is often coexistent with other comorbidities, including valvular disease, such as tricuspid regurgitation, severe pulmonary hypertension, and atrial fibrillation, with significant and independent morbidity and mortality rates [4]. The role of these co-existing factors in this setting is not well-known and should be evaluated in future research.
Studies have shown that TMVR is associated with higher rates of paravalvular leak compared to TAVR; this could be attributed to reduced anatomical support, asymmetrical annulus or asymmetric leaflets in mitral valve compared to aortic valve [4,10]. Additionally, post-dilation of mitral prosthetic valve could potentially be challenging and risky; due to the close proximity of the mitral valve to the left circumflex artery, the conduction system and the aortic valve. In addition, efforts to avoid damage to sub-valvular structures should be pursued [4,10]. Future improvement of the dedicated TMVR-specific device design should address these anatomical issues. As we discussed in the previous section, different transcatheter devices have been designed for the treatment of MR (and, in some cases, for off-label treatment of mitral stenosis). Most of the TMVR technologies are still under clinical

Author details
Lina Ya'qoub* and Marvin Eng Henry Ford Hospital, Detroit, MI, USA *Address all correspondence to: yaqoublina1989@gmail.com investigation. Thus, data about their rates of structural deterioration and durability is limited [10].
At the present time, clinical outcomes we have are based on data of mainly the first and second generation prosthetic valves. Outcomes may potentially improve with newer generation devices, improvement in the process of patient selection, operators' experience and innovations in procedural techniques [4]. Newer generation valves with repositionable and retrievable ability could be of benefit in certain patients. For example, the Lotus valve (Boston Scientific, Marlborough, Massachusetts) and Direct Flow (Direct Flow Medical Inc. Santa Rosa, California) valves have been successfully implanted in patients with severe MAC, however the outcomes of these valves should be assessed in future randomized clinical trials utilizing larger number of patients [11].
In conclusion, we have seen several advances in TMVR in the past decade with promising results. However, there remain challenges that need to be evaluated in future studies in order to optimize our procedural success, device evolution, and clinical outcomes to make this new cutting-edge technology available for high-risk patients.
© 2021 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/ by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.