The transcatheter valve technology pipeline for treatment of adult valvular heart disease

Abstract

The transcatheter valve technology pipeline has started as simple balloon valvuloplasty for the treatment of stenotic heart valves and evolved since the year 2000 to either repair or replace heart valves percutaneously with multiple devices. In this review, the present technology and its application are illuminated and a glimpse into the near future is dared from a physician's perspective.

Gaining knowledge is followed by rising doubt

J.W. Goethe (1749–1832, Weimar/Germany)

Introduction

Transcatheter valve technology (TVT) began with pulmonary valvuloplasty (1982),¹ mitral valvuloplasty (1984)^2,3 and later aortic valve valvuloplasty (1986).⁴ While mitral and pulmonary valvuloplasty have good long-term results, the high restenosis rate after aortic valvuloplasty initiated the development of percutaneous aortic valves, which began in the early 1990s,^5–7 followed by the first human implants of a pulmonary valve by Bonhoeffer et al.⁸ for the treatment of a stenosis in a pulmonary conduit and an aortic valve replacement by Cribier et al.⁹

Especially, the clinical need for a durable less invasive treatment of aortic stenosis (AS) in an aging population and the Conformite Europeenne (CE) marking of two transcatheter valves in 2007 and later on second-generation aortic valves has changed the stage in cardiology because of huge physicians and patients acceptance. Percutaneous aortic valve implantations (transcatheter aortic valve implantation) are presently followed by a bunch of devices to treat mitral regurgitation and the ongoing search for percutaneous mitral valve replacement, followed by the treatment of severe tricuspid regurgitation (TR).¹⁰ Transcatheter valve technology today has the potential to treat the whole spectrum of valvular heart disease. The present technology and future developments of transcatheter valve technology are at the scope of this article.

Transcutaneous mitral valve therapy

Treatment of mitral stenosis

Although declining in the Western population, mitral stenosis (MS) is still a frequent disease in the undeveloped and developing countries. About >15 million people are affected each year by rheumatic heart disease with 233 000 deaths and 282 000 new cases each year.¹¹

Evidence-based interventions such as surgical or percutaneous therapy in low-income countries are being performed in only ∼20% of the cases.¹² Balloon valvuloplasty is a TVT of choice which can be easily performed with the Inoue balloon catheter, a bell-shaped balloon, that can be positioned transseptal within the mitral valve area under fluroscopic guidance in local anaesthesia.²

Short-term results are favourable: mortality 0.4%, cardiac tamponade 0.2%, arterial embolism 0.6%, and severe mitral regurgitation 4.1%.¹³ Long-term results demonstrate that 30% of patients have a good functional result after 20 years.¹⁴ These long-term results of the percutaneous balloon mitral valvuloplasty (PBMV) with the Inoue balloon are so satisfying that they consequently almost eliminate surgical commissurotomy in most patients.¹⁵ Most important, however, is the adequate patient selecting according to echo parameters.¹⁶ According to Sugeng et al.¹⁷ 3D planimetrically from the left ventricular side is the most accurate method for the mitral valve area determination (Figure 1A and B). Similar precision can be obtained by MRI measurements.¹⁸

The prediction of PBMV success according to anatomic variables is mostly being classified by the Wilkins–Boston score.¹⁹ However, a new scoring system has been introduced to predict a non-favourable acute outcome (Table 1, Nunes score)²⁰: seems to be more predictive than the Wilkins–Boston score. Both in the European guidelines from 2012 and in the AHA/ACC guidelines 2014 PBMV have a prominent position.^21,22 Both guidelines recommend PBMV as primary therapeutic option in symptomatic patients with a valve area <1.5 cm², the European guidelines recommend balloon valvuloplasty also in patients with less favourable anatomy while the AHA/ACC guidelines are more restrictive in that group of patients with only a IIb recommendation.^21,22 Irrespective of the anatomic score, PBMV is contraindicated according to Table 2.²³ As stated in the European guidelines, even in patients with a higher Wilkins score, the mortality of PBMV is low (1%)^13,24,25

• – displacement of mitral valve leaflat <12 mm

• – asymmetry of the opening area >1.25

• – involvement of subvalvular apparatus

• – a valve area ≤1.0 cm²

The procedure is considered successful when an increase of mitral valve area >1.5 cm² without significant mitral regurgitation (>2/4) can be demonstrated by echo the day after the procedure. Pressure half time determinations of the transmitral gradient during the procedure is not reliable, since this parameter is very much depending on LA/LV compliance which is affected by the procedure itself.¹⁶

Transcatheter mitral valve in valve (ViV) implantation for bioprosthetic valve dysfunction, annuloplasty ring dysfunction, or calcified mitral stenosis

The percutaneous implantation of a catheter valve in a degenerated surgical tissue valve has been shown to be effective to reduce valvular leakage, gradient, and even paravalvular leakage in aortic and mitral position.^26,27 In a recent series, Cheung and co-workers could demonstrate a significant decrease of transvalvular gradient from 11.1 to 6.9 mmHg by the implantation of an Edwards Sapien XT valve via the transapical route into a degenerated bioprosthetic valve.²⁸ So ViV therapy is a valuable option in patients with either surgical ring annuloplasty or degenerated surgical bioprosthesis. The implantation of an Edwards Sapien XT valve on a Novaflex introducer can even be performed by a transseptal route with a 82% procedure success rate.²⁹ In a Registry (VIVID) ViV or valve in ring procedures in mitral position were performed with an acceptable risk (8.5% 30 days mortality); however, smaller surgical valves (label ≤25 mm) were associated with higher gradients.³⁰

Case reports demonstrate the implantation of a Sapien XT valve also into native MS.^31–33 Personal experience (H.R.F.) underlines the feasibility of transcatheter valve-in-MS implantation through left atrial access. A major challenge of direct prosthetic valve-in-stenosis-implantations is the risk of left ventricular outflow tract obstruction by displacement of the anterior mitral leaflet (Figure 2) in degenerative calcified MS. M. Guerrero et al. recently reported the results of this approach in 64 patients, observing a significant reduction of the transmitral gradient in this severely ill patient population. Mortality at 30 days was 29.7%.³⁴ Although this procedure is hemodynamically effective, refinements have to be adapted.

Future considerations for the transcatheter valve technology of mitral stenosis

Mitral balloon valvuloplasty in its present form with an Inoue balloon is safe and effective in ∼90% of cases.^35–37

However, two obstacles are involved:

1. The Inoue balloon is expensive and according to the economic burden in undeveloped countries (see above) the open chest operation is more frequently used. So a cheap, non-disposable device would help to switch more people to transcutaneous valvuloplasty.¹² Cribier et al.³⁸ developed such a device, however, received little acceptance. In the future an easy to use non-disposable mitral splitter, which might be used exclusively by echo guidance would be a great advantage.

2. Potential complications of PBMV include severe mitral regurgitation or persisting transmitral gradients, which have a particularly high risk in patients with severe and irregular calcified valves.³⁷ In this subgroup, percutaneous prosthetic mitral valves could represent a backup strategy in the future if PBMV is not successful (see below).

Transcatheter treatment of mitral regurgitation

Multiple surgical techniques are available to restore mitral valve function. These operations can also be performed by minimal invasive surgical techniques, although the value of this technique when compared with replacement was questioned recently in functional MR.³⁹ The mortality rate of surgical mitral repair was never compared in a randomized trial to conservative medical therapy in functional MR; however, heart failure symptoms are reduced after surgery in many patients. Especially in functional MR, the operative mortality is high; however in small observational studies comparing surgery with medical therapy, a clear benefit for the earlier could not be shown. Following that observation, the AHA/ACC and the European guidelines recommend surgical valve therapy in functional MR only with a IIb recommendation status.^21,22 Therefore, transcatheter techniques might be an attractive alternative in older patients particularly when LV-function is impaired.^40,41 A review of all devices under development has been published recently in this journal.⁴²

Interventional leaflet repair

By far, the most widely utilized device is the MitraClip with over 20 000 patients treated internationally.⁴⁰ This is a leaflet repair device and is the only direct leaflet repair device in clinical use. The next most commonly used device is the cardiac dimensions CARILLON. This coronary sinus annuloplasty catheter has been implanted in over 300 patients (see below) all other coronary sinus devices have been abandoned. After this, the frequency of clinical use of the various other devices falls significantly. The Mitralign system has been used in over 65 patients in a European CE approval trial. Direct annuloplasty with the Valtech Cardioband has been utilized in over 40 patients.⁴³ The field of percutaneous mitral valve replacement is new, with half a dozen devices having accomplished early human implants and at the time of this writing fewer than 60 total patients (see below).

The MitraClip system

The MitraClip system creates a double orifice mitral opening by a percutaneous approach via the femoral vein (Figures 3 and 4). The device uses a transseptal puncture to gain access to the left atrium. After careful manoeuvring across the mitral leaflets, while avoiding the chordae, the anterior and posterior leaflets is grasped at the A2–P2 segments. By closure of the device, the operator approximates the leaflet edges and creates a double orifice (Figure 5). The tissue bridge formed between the anterior and posterior leaflets reduces the MR. Several subsets of patients undergoing MitraClip therapy have been studied. The range of MitraClip studies includes treatment of surgical candidates in the EVEREST II randomized trial comparing mitral therapy with the MitraClip compared with standard surgery in good surgical candidates.⁴⁵ The predominant findings of the trial are that surgery clearly reduces the severity of MR more effectively than MitraClip, but despite this, both procedures have similar impact on clinical symptoms and left ventricular chamber volume and dimensions, and of course, the percutaneous therapy is associated with a better safety profile because blood transfer were less often being used.⁴⁵ This trial was undertaken in predominantly patients with degenerative MR (DMR). There is a robust global experience with MitraClip in high-risk patients for surgery with predominantly functional mitral regurgitation (FMR). The high-risk FMR population represents a large group of patients with heart failure and previously no effective option for therapy. The international experience in this group has demonstrated that the MitraClip procedure can be performed safely, given the high risk of this patient cohort. Typical 30-day mortality is <5% in a population with a Society for Thoracic Surgery predicted mortality between 10 and 20%, as shown in a recent meta-analysis of 21 studies including over 6000 high-risk MR patients.⁴⁶ Approval for MitraClip in the USA however, has been restricted to prohibitive risk for surgery patients with DMR.

There remains controversy regarding the utility of MitraClip in patients with FMR and heart failure. While there are numerous registry experiences that corroborate the earlier findings from the REALISM registry, there is yet to be a randomized comparison of MitraClip with optimum medical therapy in this high-risk population.⁴⁷ Some of the concern is based on the recognition that MR associated with heart failure is more of a left ventricular disease than a valvular problem. Thus, the comparison of MitraClip with medical therapy seems most appropriate (Figure 6). This comparison is being undertaken in the COAPT (Clinical Outcomes Assessment of the MitraClip Percutaneous Therapy for High Surgical Risk, www.clinicaltrials.gov, identifier NCT01626079) randomized trial, with 430 patients randomized 1:1. Major inclusion criteria include LVEF 20–50%, hospitalization for heart failure in the prior year or elevated BNP, and treatment with optimal guideline directed medical therapy including cardiac resyncronization therapy. Two similar randomized trials are being conducted in Europe, RESHAPE HF (Randomized Study of the MitraClip Device in Heart Failure Patients With Clinically Significant Functional Mitral Regurgitation, www.clinicaltrials.gov, identifier NCT01772108) and Mitra-FR Trial is also on its way (www.clinicaltrials.gov.NCT01920698).

Coronary sinus annuloplasty, direct annuloplasty, and chordal repair

The largest human clinical experience is with leaflet repair using the MitraClip. Several annuloplasty devices have also accrued some clinical use. The Cardiac Dimensions CARILLON has CE approval and is being implanted in several countries. The device is an indirect annuloplasty therapy, with a wireform implant in the coronary sinus (Figure 7A). Nine-French jugular venous access is used to implant the device in the coronary sinus, which results in cinching of the coronary sinus and thus the posterior mitral annulus. The magnitude of treatment effect is similar to what has seen with MitraClip.⁴⁸ A particular unusual feature of the response to a CARILLON device is a delayed response. Progressive reductions in mitral annular dimensions have improvements in left ventricular remodelling and clinical measures such as 6-min walk tests have been noted with a CARILLON device. Some plans for additional trial developments are being considered.

The Mitralign approach uses direct retrograde transaortic access to the left ventricle to implant pairs of pledgets in the posterior annulus near the commissures, to plicate the commissures and thus the annular circumference (Figure 7B). The Mitralign device has completed enrolment in a CE approval trial (ALIGN, www.clinicaltrial.gov: NCT01740583). Over 65 patients have been treated with this system. The results of this trial have yet to be published.⁴⁹

Direct annuloplasty with a transseptal implant that mimics a surgical partial annuloplasty ring has been accomplished with the Valtech Cardioband (Figure 7C). This approach uses transseptal access and screw anchors to implant an incomplete ring along the posterior annular surface, and then uses a cable to tighten the ring to accomplish a diminution at the annular circumference.⁴³ This procedure has been successfully performed in increasing numbers of patients in Europe and received CE mark for the treatment of functional MR in September 2015.

Chordal repair is possible by a transapical approach with the implementation of new chordea however, published clinical results are missing.⁵⁰

Transcatheter mitral valve replacement

While mitral repair devices have been under development for over a decade, percutaneous mitral valve replacement prostheses have just started to undergo human clinical use⁵¹ (Figure 8). Experience with existing transcatheter aortic valve implantation (TAVI) devices with implants in degenerated bioprosthetic surgical valves, or in heavily calcified native MS have shown the feasibility of transcatheter mitral valve replacement.^31–33 Numerous start-up companies have recognized the potential demonstrated by transcatheter aortic valve therapy to implement valve replacement in the mitral position.^42,52 This has proved challenging because not only are the patients generally more complex and sicker than seen for TAVI but also the mitral annulus is D shaped and much larger. There is no calcium to anchor the frame of the implant. A large bulk necessary to fill the larger diameter of the mitral valve has necessitated transapical or surgical transatrial implant approaches to date. The appeal of valve replacement devices is that they should completely eliminate mitral regurgitation, whereas the percutaneous repair approaches have a greater likelihood to leave residual MR. This consideration not withstanding, the likelihood that repair approaches will have some utility remains strong. The greater technical challenges in developing replacement devices mean at the very least that this will take some years.

Transcutaneous aortic valve therapy

Percutaneous aortic balloon valvolutomy

Percutaneous aortic balloon valvolutomy was first introduced into the therapeutic armamentarium in 1986 to treat compassionate patients with AS.⁴ Because of restenosis within a year, and disappointing long-term survival, PABV may presently only be considered as a bridge to surgery or TAVI in haemodynamic instable pts, or in pts with severe AS requiring an urgent non-cardiac operation.^21,53

In the PARTNER B trial >80% of those patients randomized to so-called standard treatment received a PABV however, the long-term mortality was doubled when compared with the TAVI group, indicating the low efficacy of PABV.^54,55 However, surprisingly the use rate of PABV increased by 158% within the time period 1998–99 to 2009–10 in the USA.⁵⁶ It might be speculated that this undesirably trend will be reversed with the broad entry of TAVI in the US market.

Transcatheter aortic valve implantation

Multiple papers and exhaustive reviews on TAVI including patient selection, unresolved issues, and procedural aspects have been published since the first human implant in 2002 by Alain Cribier.^9,57,58 Many valves for TVT have been launched meanwhile to the European market and are arising in the USA (Figure 9). So in this chapter, we will focus on important upcoming developments for patient selection (risk scores) and special issues such as prosthesis sizing, prosthesis technical refinements, and logistic modifications to make TAVI procedures simpler, safer, and cheaper.

Patient selection

Patient selection for TAVI includes individual risk stratification and is a matter of ongoing debate. Transcatheter aortic valve implantation should be performed in patients with significant AS to achieve either functional or prognostic benefit.^21,59 The first question therefore is to establish the correct diagnosis and determine whether AS is truly severe and related to the patients symptoms. Secondly, particularly in high-risk patients comorbidities require careful evaluation to determine whether they prevail functional or prognostic improvement after AV replacement.

According to recent guidelines, severe AS is defined as an aortic valve area <1.0 cm², a mean gradient >40 mmHg and a maximum jet velocity >4 m/s according to echocardiography. However, in many patients echocardiographic measurements are non-conclusive despite a valve area <1 cm² and a preserved ejection fraction >50%. In clinical practice, AS can further be subclassified according to the transvalvular flow and gradient^60,61 (Figure 10). Current data suggest the highest survival of medical treatment in patients with normal flow, low-gradient AS, the worst in low-flow, low-gradient AS, and intermediate survival between.⁶¹

In another investigation, patients with low gradients despite severe AS were compared by means of propensity match scores with those either treated with surgical aortic valve replacement (SAVR) or medical therapy. Medically treated patients presented a two-fold increase of all-cause 30-day mortality compared with SAVR.⁶³ In the Partners A and B trials, low transvalvular flow defined as a stroke volume index (SVI) <35 mL/m² was identified as an independent predictor of early mortality in all patients with AS, whereas ejection fraction and transvalvular gradient were not.⁶²

So in severe AS, a reduced SVI <35 mL/m² seems to be a significant predictor of early mortality with medical therapy that can be improved by valve replacement irrespective of ejection fraction. However, patients treated with TAVI and reduced ejection fraction and reduced stroke volume have the highest mortality compared with those with low flow and preserved ejection fraction.⁶⁴

The question when a patient is too sick to undergo TAVI touches ethical issues. The guidelines recommend a medical strategy in patients with a life expectancy <12 months due to non-cardiac comorbidities.²¹ Although this time period is a rather arbitrary cut-off value it is frequently followed in clinical practice. From the Partner B trial, we know that patients with an STS score ≥15% undergoing TAVI had no survival benefit over patients treated medically within the first 2 years.⁶⁵ Therefore, in this so-called Partner C cohort with STS score ≥15% therapy should be limited to medical treatment.

Risk prediction in patients undergoing transcatheter aortic valve implantation

The Partner A and CoreValve high-risk trial demonstrate that TAVI is equivalent or even superior to SAVR in high-risk patients.^66–68 However, risk scores being used in these trials (STS score) were not derived predominantly for risk prediction in valvular heart disease. Better risk scores predictors have been developed (Euro Score II, STS valvular score.)^69,70 The most advanced score for risk prediction currently is the German Aortic Valve score, which uses 21 variables derived exclusively from >8000 aortic valve procedures (AKL Score).⁷¹

However, these risk predictors are still imprecise. For decision management in the heart team and for patient information, an exclusive score dedicated to calculate peri-interventional TAVI risk is required. Apart from clinical variables procedural aspects such as access site, annulus size, and calcification have to be included for reliable risk prediction. Specific scores dedicated to TAVI predicting 30 days mortality have meanwhile been developed: the FRANCE 2 risk score and the OBSERVANT score.^72,73

Sizing

Echocardiography has been widely used to evaluate annular dimensions and assist selection of an appropriately sized THV. However, it is increasingly clear that the non-circular ellipse that is the normal annulus require 3D imaging for optimal evaluation. 3D CT imaging has rapidly become the current standard of care, while 3D transoesophageal echocardiography may offer similar and additional benefits.^74,75 Mean diameter can be utilized to give a general sense of the annular dimensions. However, perimeter and area are currently favoured. Perimeter has the theoretical advantage of being less affected by annular asymmetry, but the disadvantage of being more dependent upon smoothing algorithms.

Access route

Early transarterial delivery systems required 24French sheaths with external diameters of over 8 mm.^76,77 With the limited experience of early operators, vascular complications were frequent, and frequently associated with bleeding, transfusions, morbidity, and mortality. Greater operator experience and routine access evaluation with CT or angiography reduced the frequency, and lessened the consequences of, vascular complications.

Femoral access, with superior clinical outcomes, is currently the dominant access route. For patients with contraindications to femoral arterial access, the currently favoured access alternatives are apical, aortic, and axillary. Carotid, transeptal, and caval-aortic access are generally reserved for special situations. The decision between these access routes depends on anatomic and other patient-specific factors, compatibility with the specific THV selected, and operator experience.

However, arterial access technology and techniques continue to improve; the current CoreValve Evolut R and Sapien 3 delivery systems are now 5–6 mm in diameter and vascular complications continue to fall.^78,79 The large majority of patients are now candidates for fully percutaneous transarterial access and closure.

In a recently published PARTNER Trial substudy risk adjustment by 111 preprocedural variables was performed and 501 well-matched transfemoral and transapical TAVI patients were compared by using propensity-score matching. Matched transapical patients experienced more adverse procedural events, in particular 19 vs. 12% mortality within the first 6 months and slower recovery.⁸⁰

Current device selection

Much has been made of the comparisons between early generation balloon- and self-expandable THVs. Both can achieve excellent outcomes, but both have deficiencies. As this review comes to publication, the two dominant platforms are the Sapien 3 and CoreValve Evolut R. Both have low profiles of 14–16French.^78,79,81 Neither will require pre-dilation. Accurate implantation will be relatively routine with the sophisticated Commander positioning, or Enveo repositioning, systems. The Sapien 3 with its external seal may have fewer important paravalvular leaks and new sizing guidelines may reduce concerns about annular injury. This balloon-expandable valve seems also be more applicable than self-expandable valves in bicuspide AS. The need for pacemakers may be less frequent with the CoreValve Evolut R than the previous generation device. The supra-annular leaflets of the CoreValve Evolut R will result in lower gradients in small annuli. The debate will continue. Cost differentials may become increasingly a factor in device selection; but to a large degree physician familiarity and preference will be the most important differentiator.

Many newer valves have recently become available (Figure 9). These have largely been designed with the goals of improving ease and accuracy of implantation or reducing paravalvular leaks. The mechanically expandable Lotus valve and the inflatable Direct Flow valve can both achieve excellent sealing and be fully deployed and then fully recaptured.^82,83 The former suffers from high pacemaker rates and the latter from higher gradients. The Acurate valve has some features that facilitate accurate positioning.⁸⁴ The self-expanding JenaValve and Engager valves incorporate clipping mechanisms that facilitate positioning and fixation; particularly desirable in patients with non-calcified valves or aortic regurgitation.^85,86 However, the Engager valve was recently withdrawn from the market. For aortic regurgitation, a clipping mechanism on the native leaflet or otherwise active fixation is mandatory because without in calcified stenosis the friction is little. In addition due to large annuli in this group of patients bigger valves have to complete the armentarium. JenaValve is presently the only valve that is CE-marked also for treatment of pure aortic regurgitation, with the concomitant advantage of anatomic positioning, that means leaflet position resembles the native leaflet positioning.⁸⁵

The repositionable Portico valve has similarities to the Evolut R, although direct comparison is controversial.⁸⁷ The self-expanding Centera valve also has features that assist positioning and sealing, a unique motor-driven loading and release system, and is 14French sheath compatible.⁸⁸ For the most part however, these newer valves suffer from larger delivery profiles, more complex deployment, as well as a lesser body of clinical experience, although catch up may be rapid.

Durability

Manufacturers routinely demonstrate in vitro durability using automated haemodynamic testers under a variety of conditions (e.g. variable pressures and incomplete expansion) for a minimum equivalent of 5 years. Multiple clinical studies have documented mid-term durability of THVs. Five-year durability has been reported for the Sapien, Sapien XT, and CoreValve implants.^55,68,89,90 Follow-up beyond 5 years remains anecdotal.⁹¹ For the most part leaflet degeneration has been mild, with only scattered reports of structural degeneration, primarily due to calcific stenosis.^92–94

What will happen when TAVI is applied to lower risk patients who may live long enough for their transcatheter valves to fail? Surgery may be an option for some. However, favourable experience with THV implantation in failed surgical and transcatheter bioprostheses has demonstrated that TAVI may more easily repeated than open surgery.^95,96

Paravalvular regurgitation

With the initial experience with TAVI, paravalvular leaks were omnipresent and frequently clinically important. Fortunately, with an improved understanding of how to optimally size and position THVs severe leaks have become relatively infrequent.^76,77 Unfortunately, moderate and mild leaks remain common. It does appear that moderate leaks can adversely impact functional benefit and late survival. However, while mild leaks can be predictive of poorer outcomes in some analyses, it is questioned whether this is a true cause and effect relationship or an artefact.⁹⁷ Quantification of paravalvular leaks is controversial and varies markedly according to the imaging modality, effort, and core laboratory criteria applied.⁹⁸ Regardless, newer valves that are repositionable or have improved seals have been shown to markedly reduce leaks.^79,82

Stroke

Embolization of calcific material as a consequence of traumatic injury to the diseased valve leaflets appears to be the dominant mechanism of stroke.⁹⁹ Additional mechanisms include atheroembolism from the ascending aorta, cerebral vessel injury, and importantly thromboembolism (particularly in association with atrial fibrillation).¹⁰⁰

The early TAVI experience was marred by a relatively high risk of procedural stroke; largely as a consequence of early crude valve delivery catheters, a steep learning curve, and a substrate of severely ill patients with multiple comorbidities.^76,77 Reducing the risk of cerebral embolism has been achieved with less traumatic delivery catheters, improved operator technique, minimizing manipulation (valvuloplasty, repositioning) of the diseased valve, and a trend towards patients at lower risk of stroke.

The feasibility of cerebral embolic protection has been demonstrated with several filter-like devices; whether embolic protection is truly efficacious is currently under investigation.¹⁰⁰ The prevention of stroke also relies on the optimization of antithrombotic therapy after TAVI, which remains empirical. This is of particular importance, given the relationship between post TAVI atrial fibrillation and stroke. Regardless, recent experience has shown that stroke rates can be achieved that are relatively comparable with, or probably lower than, open surgery.^66–68,101

Atrioventricular block

Atrioventricular block is a known complication of aortic valve replacement.¹⁰² Balloon-expandable THVs have been associated with a frequency of high-grade block requiring implantation of a permanent pacemaker comparable with that associated with surgery. In contrast, conduction block has been considerably more frequent with the self-expandable CoreValve device and requiring a longer period of monitoring with in-dwelling temporary pacemakers, much higher pacemaker rates (Figure 11). High rates of new left bundle branch block have also been a concern, although the implications of this are controversial.^103,104

Evidence suggests a reduction in pacemaker rates with CoreValve can be achieved if implantation deep into the left ventricular outflow tract can be avoided. Similarly, lower pacemaker rates have been reported with other self-expanding valves that are less likely to extend into the outflow tract. In contrast, high rates have also been reported with the mechanically expanded Lotus valve.

Mortality

The PARTNER trial demonstrated that TAVI could dramatically improve survival and quality of life in patients declined surgical valve replacement.^55,76,77 As a result this is now generally accepted as the default therapy for patients in whom the risk of surgery is prohibitive. Shortly afterwards the PARTNER 1A trial demonstrated TAVI survival equivalent to surgery in high-risk surgical candidates, with equivalence durable at 5-year follow-up.^55,68,77 More recently, studying a ‘less high-risk’ cohort, the US CoreValve pivotal trial demonstrated survival arguably superior to surgery. And even more recently, the PARTNER Sapien 3 trial documented an extremely low 30-day mortality of 2.2% in high-risk patients and a game changing 1.1% mortality in intermediate risk patients (lower than the STS-predicted mortality of 8.6 and 5.3%, respectively).^66,67 These and multiple other lines of evidence have led some to suggest that TAVI may have the potential for a lower procedural mortality than surgery in all risk groups.

Future developments

Future developments are aiming to make the TAVI procedure safer, easier, and more reliable. It appears likely that in the near future we will see a number of balloon-, self-, and mechanically expandable systems that facilitate routine and safe arterial access, accurate positioning, effective sealing, excellent haemodynamics, and adequate durability. While some of these systems will not find favour, it appears unlikely that any one system will clearly dominate over the long term. However, it seems likely that TAVI in general will become the dominant therapy for AS. Presently, the hospital 30 days mortality rate is at ∼5% (see above) with an ∼2.5% stroke rate, pacemaker rate 10–30%, and vascular complications at a rate of ∼5% outside randomized trials in Registries.¹⁰⁵

To avoid all of these risks and side effects, technical developments need to address the shortcomings of current devices. Technical refinements should address the operator needs which are: Unfortunately however, we need to prioritize the features of new TAVI devices to achieve progress.

1. Small flexible introducers to allow all procedures to be performed percutaneously from the femoral arteries

2. Easy and precise anatomical positioning in all fluoroscopic planes (no need for orthograde predefined projections)

3. No need for rapid pacing during implantation

4. No induction of AV blocks

5. No gradients after implantation

6. No regurgitant flow

7. No strokes

8. Valves for annulus sizes >30 mm

9. Treatment of aortic regurgitation

10. Lifelong durability?

Lifetime durability cannot be reached with any biological heart valve, especially in young age groups. Until now there is no evidence that TAVI has less durability when compared with surgical tissue valves. Ongoing developments concentrate on none biological membranes such as UCL-valve, these nitinol membranes will be thinner than biological material and allow smaller introducers.¹⁰⁶

Another way for smaller introducers would be to cryo dry the biological tissue and to preload the valve (Colibri valve) or to assemble the valve inside the aorta from two or more pieces (Valve Medical) (Figure 12).¹⁰⁷

The feeler or clipping technology of nitinol stent frames allows easy and precise anatomical positioning, thus deployment is reliable after the first attempt at the right position. Because frequent repositioning might increases the perioperative risk of stroke minimal irritation during deployment with that technology might reduce the stroke rate. In addition, septum irritation is reduced and primary aortic regurgitation without calcification can be addressed without the risk of valve embolization.¹⁰⁸

However, the introducers with this technology are bigger due to a more complex deployment step and a bigger stent frame. Conceptual this is the easiest technology with many advances. Paravalvular regurgitation can be avoided by abluminal sealing shirts presently realized by the Lotus and Sapien 3 valves. Sealing can also be obtained by gel shirts, expanding after implantation. The later could be charged with antimicrobial substances, used for local therapy of endocarditis. Elastic and soft shirts around the basis of the stent frame would also reduce septum irritation and the frequency of AV blocks.

In the future, TAVI might replace SAVR with biological valves completely. Valve improvements will make valve implantation easy, quick, and safe without general anaesthesia and allow a single operator.

Transcatheter treatment options for tricuspid regurgitation

In contrast to interventions on the aortic and mitral valve, the tricuspid valve has received less attention from interventional cardiologists, cardiac surgeons, and researchers. Except for balloon valvuloplasty in tricuspid stenosis no specific interventional concepts for tricuspid valve disease is currently established.^109,110 For TR, limited experimental data have been published and compassionate treatment in isolated human cases has been performed with different approaches.^111–113

There is a large unmet need for a transcatheter treatment of TR. Tricuspid regurgitation is common in patients with late stages of left heart valve- or myocardial disease, treatment is therefore increasingly observed. In the majority of patients, TR is not related to primary valve pathology but rather functional. Moderate or severe TR has a significant impact on functional status and long-term survival and has been recognized as an independent risk factor for poor long-term survival.^114–117 In these often high-risk patients in an advanced disease state, tricuspid surgery carries an operative mortality of up to 22% and is therefore frequently not performed.¹¹⁸

Transcatheter repair of the tricuspid valve

Size and flexibility of the tricuspid annulus and the surrounding myocardium as well as the loss of anatomic landmarks under pathologic conditions counteract the positioning and long-term fixation of devices for tricuspid repair or replacement.

In analogy to mitral valve repair, transcatheter concepts of annular plication have successfully been applied on the tricuspid valve to achieve circumferential annulus reduction. The Mitralign Percutaneous Annuloplasty System (Mitralign, MA, USA) has recently been used successfully on the tricuspid valve for compassionate human treatment.¹¹³ In a transcatheter approach, 2–4 pledged-enforced anchores placed through the tricuspid annulus and plicated together, resulting in annuloplasty and bicuspidization with reduction of TR. Although technically challenging, this concept has great potential for treatment of TR as it can be performed irrespective of the annulus size and adjacent anatomic structures, no implant is left behind and even a ‘partial’ correction of TR is possible, thus avoiding the haemodynamic burden of acute complete correction of TR to the right ventricle. Whether this technique will achieve durable results will have to be investigated in clinical studies.

The TriCinch System™ (4Tech Cardio Ltd) is another technology aiming to reduce the degree of TR. By transcatheter placement of a screwed anchor in the septal portion of the TV annulus, tension is applied and the septolateral diameter of the annulus and subsequently the degree of regurgitation is reduced. The tension is maintained by wire-connecting of the screw to a self-expandable stent anchored in the inferior vena cava. The device is currently investigated in a phase 1 clinical trial.¹¹⁹

Annuloplasty might also be performed by means of percutaneous tricuspid ring implantation. One investigational device specifically developed for TV repair is the Millipede annular ring (Millipede, LLC, Ann Arbor, MI, USA). This device mimics surgical annuloplasty by transcatheter implantation of an expandable and contractable ring that uses a novel attachment technique with many small barbed anchors to secure it in place. Prior to ring fixation, the annulus is expanded by a dilator to size and shape of the ring.¹¹⁹ The ring is then anchored to the annulus and contracts to a predefined size, resulting in size reduction of the annulus.

Another concept for TR treatment involves the positioning of a transvalvular spacer across the TV as platform for native leaflet coaptation. This approach was successfully implemented first-in-man using the Edwards FORMA Repair system, which was implanted though the subclavian vein and anchored in the RV apex.¹²⁰

Transcatheter replacement of the tricuspid valve

Transcatheter valve repair is conceptually attractive, however, due to technical challenges, it is uncertain whether any of the above concepts can be successfully implemented. Considering the large unmet need for an interventional option for inoperable patients, percutaneous valve implantation could be a feasible alternative. From the interventional perspective, there are two basic approaches depending on the anatomic site of prosthetic valve implantation—an orthotopic vs. heterotopic valve replacement.

In orthotopic valve replacement, the prosthetic valve is deployed at the level of the TV annulus between the right ventricle and right atrium. This approach was investigated by Boudjemline et al.¹²¹ by means of implanting a double-disc nitinol stent with a semilunar valve into the tricuspid annulus. Due to the anatomic structure and the flexibility of the surrounding myocardium, this site of implantation offers little resistance for orthotopic long-term fixation of stent-based valves with the current technique. In functional TR, annulus dilatation may reach >70 mm associated with a loss of anatomic landmarks between the RV and RA. A device intended for orthotopic TV replacement would therefore require unique solutions for stent- and catheter design as well as tissue valve engineering.

In human patients, orthotopic tricuspid valve replacement has been used with promising results only as valve-in-valve and valve-in-ring procedure using either the balloon-expandable Edwards Sapien or the Melody valve.^122–125 As with most low-volume procedures, there is no published data from larger patient series regarding the outcome and long-term function of these devices in tricuspid position. Nonetheless, it is probably correct to predict that the percutaneous approach will be the treatment of choice in patients requiring a reintervention on the tricuspid valve after previous tricuspid surgery.

An feasible alternative to orthotopic approaches is caval valve implantation (CAVI), which involves the implantation of stent valves in a heterotopic position into the inferior and superior vena cava. This concept has been investigated preclinically with encouraging results and meanwhile been applied for compassionate pts.^{10,111,112,126} The CAVI procedure is technically simple and can be rapidly performed without interference to cardiac structures. Devices do not interfere with any preexisting transtricuspid pacemaker or defibrillator leads, which might represent a limitation for orthotopic procedures on the tricuspid valve. The persistence of right atrial volume overload, the ventricularization of the right atrium and the increase of RV afterload are potential limitations of the procedure and its long-term impact on RA and RV function is currently unknown. Post-implant device function has been observed up to 24 months following implantation (Figure 13).

However, pulsatile blood flow and systolic flow reversal in the caval veins are prerequisites for the proper function of the caval valves and haemodynamic proof of TR is required prior to heterotopic implantation (Figure 14).

Transcatheter valve technology for the treatment of pulmonary valve disease

Pulmonary valve disease is rare in the adult population, either following Fallots surgery in the childhood, degeneration of biological valves following Ross surgery for the treatment of AS or undetected congenital pulmonary disease.

It is generally accepted that pulmonary balloon valvuloplasty is indicated when the transpulmonary gradient exceeds >50 mmHg.¹²⁷ It is a straight forward procedure with stepwise balloon diameter increase not exceeding 1.4:1 of the pulmonary artery diameter, until the gradient is <30 mmHg.

The long-term results are excellent and restenosis is unlikely; however a balloon to artery ratio should not exceed 1.4:1 otherwise in the long run pulmonary regurgitation is at risk.^128,129

Pulmonary regurgitation and RV enlargement appears to have consequences for left ventricular function. The exact mechanisms are unclear, but diastolic ventricular interaction along with prolonged ORS duration may be involved.^128,130 Right ventricular stretch and dilation creates a mechanoelectrical substrate for reentry circuits followed by sustained ventricular tachycardia and sudden death. Pulmonary regurgitation should be treated, if RV volume index exceeds 150 mL/m².¹³¹

In degenerated outflow conduits or biological valves restenosis of solely balloon dilatation of pulmonary stenosis (TPBV) is followed by frequent restenosis, therefore in this setting the implantation of percutaneous valves is mandatory.

Presently, the Melody valve and the Sapien XT valves are used for the treatment of pulmonary stenotic biological valves or grafts or in pts with significant pulmonary regurgitation.^132,133 Both valves should be deployed into a covered or uncovered stent frame, which enables easy positioning of the valve and prevents from fractures in the valve stent body. Before deployment a coronary angiography is mandatory to outline the anatomical interaction between the pulmonary valve and the left main coronary ostium to prevent obstruction of the latter during the deployment of the valve.

Presently, the treatment of pulmonary regurgitation is restricted to the maximal size of the available stents (MELODY valve ≤20 mm, SAPIEN ≤27 mm); therefore, self-expanding valves for the treatment of pulmonary regurgitation are under development.

Epilogue

Open chest valvular treatment is presently followed by closed chest transvascular treatment. What follows next? We are at the beginning to understand the genetic background of AS,¹³⁴ but the understanding of environmental factors, fluid dynamics, and cellular factors contributing to valvular heart disease is beyond the scope of our present understanding. Considering these aspects, the evolution of valvular treatment is still unsatisfying. Better understanding of the pathophysiology of valvular tissue destruction has to be undertaken by cellular scientists to develop cellular targets to prevent rather than to treat valvular heart disease.

Authors' contributions

A.L. accessed the data to verify the manuscript's scientific integrity, drafted and critically revised the manuscript and gave final approval of the submitted manuscript. J.G.W. conceived and designed the research, accessed the data to verify the manuscript's scientific integrity, drafted the manuscript, made critical revision of the manuscript for key intellectual content, gave final approval of the submitted manuscript, and take responsibility for the entire content. T.F. accessed the data to verify the manuscript's scientific integrity, made critical revision of the manuscript for key intellectual content, gave final approval of the submitted manuscript, and take responsibility for the entire content. H.R.F. drafted the manuscript, made critical revision of the manuscript for key intellectual content, accessed the data to verify the manuscript's scientific integrity, gave final approval of the submitted manuscript, responsible for the entire content.

Conflict of interest: T.F. reports grants and personal fees from Abbott, grants and personal fees from BSC, grants and personal fees from Edwards, during the conduct of the study. J.G.W. reports grants and personal fees from Edwards lifesciences, outside the submitted work. H.R.F. reports personal fees from consulting fees by JenaValve, outside the submitted work and is co-founder of JenaValve; personal fees from consulting fees by P&F, outside the submitted work. A.L. reports personal fees by consulting fees by P&F, outside the submitted work.

References