Summary
Tricuspid valve disease is mainly represented by tricuspid regurgitation (TR), which is a predictor of poor outcome. TR is usually secondary, caused by right ventricle pressure or volume overload, the leading cause being left-sided heart valve diseases. Tricuspid surgery for severe TR is recommended during left valve surgery, and consists of either a valve replacement or, most often, a tricuspid repair with or without prosthetic annuloplasty. When TR persists or worsens after left valvular surgery, redo isolated tricuspid surgery is associated with high mortality. In addition, a sizeable proportion of patients present with tricuspid surgery deterioration over time, and need a reintervention, which is associated with high morbi-mortality rates. In this context, and given the recent major breakthrough in the percutaneous treatment of aortic and mitral valve diseases, the tricuspid valve appears an appealing challenge, although it raises specific issues. The first applications of transcatheter techniques for tricuspid valve disease were valve-in-valve and valve-in-ring implantation for degenerated bioprosthesis or ring annuloplasty. Some concerns remain regarding prosthesis sizing, rapid ventricular pacing and the best approach, but these procedures appear to be safe and effective. More recently, bicuspidization using a transcatheter approach for the treatment of native tricuspid valve has been published, in two patients. Finally, other devices are in preclinical development.
Résumé
La pathologie tricuspide est le plus souvent une insuffisance tricuspide (IT), qui est un facteur de mauvais pronostic. L’IT est en général secondaire à une surcharge de pression ou de volume du ventricule droit, la première cause en étant une valvulopathie gauche. Durant une chirurgie valvulaire gauche, la correction d’une IT sévère est recommandée et consiste en un remplacement valvulaire tricuspide ou en une plastie tricuspide avec ou sans anneau prothétique. Si l’IT persiste après une opération du cœur gauche, une chirurgie redux dédiée est grevée d’une lourde mortalité. Par ailleurs, une proportion non négligeable de patients présentent une détérioration de leur chirurgie tricuspide au cours du temps et requièrent une réintervention, associée à une morbi-mortalité élevée. Au vu des récentes avancées des techniques percutanées concernant les valves aortiques et mitrales, la valve tricuspide apparaît comme un challenge défi majeur, malgré des obstacles propres. Les premières applications ont été les techniques percutanées de « valve-in-valve » et de « valve-in-ring » pour les dégénérescences de bioprothèse et d’anneau tricuspides. Même si des questions persistent concernant l’évaluation optimale de la taille de prothèse à implanter, la stimulation ventriculaire rapide ou l’approche la plus appropriée, ces procédures apparaissent sûres et efficaces. Plus récemment, la bicuspidisation percutanée a été publiée chez 2 patients pour le traitement d’une IT sur valve native. Enfin, d’autres procédés sont en développement au stade préclinique.
Background
Dysfunction of a native or previously operated tricuspid valve encompasses a variety of situations, which are frequently sources of difficulty in patient management. Tricuspid disease is most often tricuspid regurgitation (TR), which is mainly secondary to left-sided heart valve disease . Tricuspid valve disease has long been ignored, with the belief that TR would improve after surgical correction of left valve disease. A large body of evidence now supports the negative effect of significant TR, and this recognition has led to more frequent indications for combined tricuspid surgery, with the inherent risk of subsequent dysfunction of tricuspid repair or replacement . Whatever the clinical context, redo tricuspid valve surgery is often associated with high morbi-mortality rates. Following the recent development of transcatheter therapies for aortic and mitral valve diseases, the possibility of lower-risk tricuspid valve intervention is therefore particularly attractive.
This review presents current status and perspectives regarding transcatheter therapies for tricuspid valve disease, both for bioprosthesis or ring annuloplasty failure and for native valves.
Rationale for percutaneous treatment of tricuspid valve disease
Native tricuspid valve disease can be either stenotic or regurgitant. The extremely rare stenotic lesions, caused mainly by rheumatic fever or carcinoid syndrome, will not be addressed in this paper. The most common disease of the tricuspid valve is regurgitation, which is more often secondary rather than caused by a primary valve lesion, particularly in Western countries . Secondary TR is caused by annular dilatation and increased tricuspid leaflet tethering in relation to right ventricular pressure and/or volume overload . Pressure overload is most often caused by pulmonary hypertension resulting from left-sided heart disease or, less frequently, pulmonary disease, while right ventricular volume overload is observed in left-to-right shunts or intrinsic disease of the right ventricle.
TR has a strong negative effect on outcomes . In a study of more than 5200 patients followed for longer than 5 years , both moderate and severe TR were associated with increased mortality. Moderate TR may indeed worsen during long-term follow-up after surgery for left-sided valvular disease, and is also associated with decreased survival . Similarly, residual significant TR after mitral surgery was responsible for a four-fold increase in late mortality in a recent study . However, management of these patients with isolated severe functional TR after previous left-sided valvular surgery is difficult because of the inherent risk of redo interventions. Indeed, operative mortality was 15% in a series of 82 patients undergoing tricuspid surgery, of whom 73% had had previous left-sided heart surgery .
Tricuspid annuloplasty is effective in reducing the severity of functional TR, improves right ventricle geometry and function, and has a favourable effect on mortality . Tricuspid valve replacement is usually limited to anatomy unsuitable for repair. Guidelines recommend the consideration of wide indications for combined tricuspid surgery in patients undergoing left-sided heart valve surgery . However, a number of patients now present with late prosthetic dysfunction after tricuspid repair or replacement. After previous tricuspid valve surgery, recurrence of moderate or severe TR may be as high as 60% at 5 years , and reoperation is necessary in approximately 20% of patients within 10 years after tricuspid valve surgery . While redo surgery is the treatment of choice for a degenerated bioprosthesis (BP) or deterioration of ring annuloplasty (RA), it may be associated with a very high mortality rate , reaching 35% at 30 days , particularly in patients with co-morbidities .
Uncertainties concerning the risk–benefit ratio of surgery in severe functional TR or dysfunction after tricuspid surgery therefore justify the search for less invasive approaches.
Since its first steps , the efficacy of transaortic valve implantation has led over time to transition to valve-in-valve implantation for a failing aortic BP and, more recently, for mitral BP and RA deterioration, with favourable immediate and mid-term outcomes . Tricuspid valve-in-valve and valve-in-ring implantation may therefore present a new therapeutic option. More recently, transcather repair techniques have been described to correct functional TR in native valves.
Tricuspid bioprosthesis, ring annuloplasty and percutaneous interventions
Transcatheter heart valve (THV) implantation has recently been reported for the treatment of a degenerated tricuspid BP or failure of RA .
Whereas the technique for valve-in-valve shares features with procedures for aortic or mitral valves, the valve-in-ring implantation presents specific challenges because of the tricuspid position .
Tricuspid bioprosthesis failure and valve-in-valve implantation
Prosthesis measurements
One of the main difficulties associated with the procedure is the proper evaluation of prosthesis size. For tricuspid BP, as for aortic or mitral valve-in-valve, the prosthesis size may be derived from an integrated approach, taking into account the manufacturer’s inner diameters and the mean diameter determined by computed tomography (CT), three-dimensional transoesophageal echocardiography (TEE) and fluoroscopy. Whenever possible, the three available techniques should be used systematically in the absence of recommendations for the assessment of prosthesis size ( Fig. 1 ). However, given the round shape of the prostheses, and based on the experience of the centre, a reliable CT scan measurement may be sufficient. Based on our experience, fluoroscopy usually provides the largest dimensions while TEE gives the smallest ones .
Transcatheter heart valve implantation procedures
The routes reported so far for tricuspid implantations have been mainly transatrial or transjugular accesses . However, the transfemoral route has also provided good results, confirming the technical feasibility of this approach . The avoidance of a thoracotomy, with its potential complications, is deemed desirable given the high-risk profile of the candidates regarding such techniques. In the published transfemoral approach series of the Bichat centre, no technical difficulty was observed .
In our institution, transfemoral procedures are performed under general anaesthesia and TEE guidance. After crossing the tricuspid valve, a J-shaped guidewire is placed at the apex of the right ventricle ( Fig. 2 ); then, an Edwards SAPIEN™ XT valve (and more recently the SAPIEN™ 3 valve) (Edwards Lifesciences, Irvine, CA, USA), mounted upside down on the catheter, is deployed by slow balloon inflation under rapid ventricular pacing (RVP) (140–180 beats per minute). RVP is performed using either a temporary lead in the left ventricle or in the coronary sinus, or using a permanent pacemaker implanted before the procedure ( Fig. 3 ). Predilatation of the BP should be avoided as often as possible to prevent the risk of cusp rupture or embolization, although their consequences are less severe than for left-sided valves. The results are assessed in the catheterization laboratory by haemodynamics, TEE and fluoroscopy, then a CT scan with three-dimensional reconstruction is performed before discharge ( Fig. 4 ).
Choice of valve: Melody ® or Edwards SAPIEN™?
THV implantation in a degenerated tricuspid BP has been reported using two different valves: the Melody ® valve (Medtronic, Minneapolis, MN, USA) and the Edwards SAPIEN™ valve.
A series of 15 Melody ® valve implantations in bioprosthetic tricuspid valves in patients with a median age of 31.5 years, most of whom had congenital heart disease, reported good results . The valve was successfully placed in all patients and the median follow-up was 4 months. Complications were one death in a patient with preprocedural multiorgan failure, one third-degree atrioventricular block requiring pacemaker implantation and one case of endocarditis 2 months after implantation. However, recent concerns have been raised regarding the increased risk of infective endocarditis in the Melody ® valve, although mainly in the pulmonary position .
A number of cases and series have also been published using the Edwards SAPIEN™ valve . A recent review of the literature reported eight case reports using the Edwards SAPIEN™ valve in patients aged 8–74 years. All procedures had good results without complications, and the valve performance after implantation was good, similar to the results of the series with Melody ® valve implantations , but the mean reported follow-up was only 3 months.
A recent series of 16 patients implanted between 2008 and 2014 has been published with both the Melody ® valve ( n = 7) and the Edwards SAPIEN™ valves (SAPIEN™ XT 26, n = 4; SAPIEN™ XT 29, n = 6) . The procedural success reached 100% in this series of young patients (median age 31 years), most of whom had congenital heart disease (14 patients) and only two of whom had an initially acquired tricuspid dysfunction.
In Bichat hospital, we implanted four patients with a BP failure, all of whom had acquired tricuspid dysfunction (three of rheumatic origin and one endocarditis) (unpublished results). This was a very different population compared to patients with congenital heart disease, with a mean age of 56 years, and all patients were implanted with an Edwards SAPIEN™ XT valve (26 mm, n = 2; 29 mm, n = 2). Of these four patients, two were already tridux, one was already redux and the last patient had a history of drug use and infective endocarditis. For the degenerated BP, two had stenotic impairment, and two had both regurgitant and stenotic BP failure. The procedural success was 100% with no conversion to surgery. During in-hospital follow-up, two complications occurred: one major gastrointestinal bleeding related to oesophageal ulceration and one severe vascular complication needing surgical management as a result of venous subcutaneous bleeding at the puncture site. No other complication occurred at 30-day follow-up.
Both the Melody ® valve and the Edwards SAPIEN™ XT valve seem to give good results for THV implantation in the tricuspid position. However, given the small sizes available for the Melody ® valve (maximal diameter of 22 mm), it may be less adequate for adult patients with acquired valve disease. In our series, for example, the sizes of the failed BPs were between 25 and 31 mm, which is not compatible with Melody ® valve implantation. Moreover, larger studies with a longer follow-up need to be conducted to assess whether the risk of infective endocarditis is really more important with the Melody ® valve.
Rapid ventricular pacing: specificities for tricuspid transcatheter heart valve implantation
For THV implantation in the tricuspid position, RVP should not be performed via a temporary right ventricular pacing lead, should the lead be jailed or damaged during the procedure. Several other strategies may be proposed .
For patients with a previous permanent pacemaker, RVP may be performed using the latter. Implantation of an additional lead in the coronary sinus may also be discussed to prevent any adverse event.
When the patients do not have a pacemaker, the different options for RVP are either a temporary lead in the coronary sinus or in the left ventricle, or direct pacing using the wire at the apex of the right ventricle and crocodile forceps.
In some particular cases, when the risk of pacemaker implantation is very high, a permanent epicardial pacemaker may be proposed before the procedure. These options are feasible without major difficulties and the choice must be individualized to patients’ characteristics ( Fig. 3 ).
Annuloplasty rings failure: specificities of valve-in-ring implantation
All the challenges previously described for valve-in-valve implantation also apply to valve-in-ring procedures. Additionally, tricupid annuloplasty rings present some specific difficulties.
Annulus measurements
Prosthesis sizing is particularly difficult for prosthetic tricuspid rings, which are non-circular, as illustrated in Figs. 1 and 4 , and surrounded by valvular tissue that cannot be detected by CT scan or fluoroscopy. In this setting, it is therefore mandatory to systematically use the three available techniques (CT scan, TEE and fluoroscopy) to determine the most appropriate size of prosthesis by using an integrated approach. The aim is to implant the valve with the diameter that is closest to the mean inner diameter of the ring, to avoid over- or under-valve expansion.
Moreover, careful balloon sizing at the beginning of the procedure may be used to help determine the annulus dimension more precisely .
Valve-in-ring implantations: mind the gap!
Only three case reports have been published so far with the Edwards SAPIEN™ valve, using the transatrial and transfemoral approaches.
A recent series of three valve-in-ring implantations with an Edwards SAPIEN™ XT valve through the transfemoral approach has been published . The sizes of the failed RAs were between 30 and 32 mm, allowing for implantation of three 26 mm Edwards SAPIEN™ XT stents. The authors reported one patient with a moderate-to-severe residual regurgitation immediately after the procedure, in a rigid and incomplete ring (Carpentier-Edwards Classic Ring; Edwards Lifesciences, Irvine, CA, USA). The regurgitation was caused by a paravalvular leak at the level of the open portion of the ring. Another patient had mild post-procedural paravalvular regurgitation, in the same type of rigid and open ring.
In a recent case report, a 22 mm Melody ® valve was implanted in a 26 mm failed RA in a 21-year-old young women . A severe paravalvular leak occurred in the region where the tricuspid ring was incomplete. This case also concerned a Carpentier-Edwards Classic tricuspid annuloplasty ring. The authors report the successful closure of the paravalvular leak using an AMPLATZER™ Vascular Plug II (St. Jude Medical, St. Paul, MN, USA), with only a mild residual leak as the final result .
The presence of a prosthetic ring has the advantage of providing the fluoroscopic landmarks and necessary anchoring for a percutaneous valve, but also has the drawback of creating a non-circular landing zone, with the inability to completely seal the open segment with the THV, and inducing underexpansion of the transcatheter valve in the small diameter of the ring. These findings are more likely to appear with the Carpentier-Edwards Classic tricuspid rings, which are rigid, making it difficult to circularize this type of ring ( Fig. 4 ).
Residual TR is, however, probably less clinically relevant than in the mitral or aortic position. Indeed, we reported a sustained improvement in the functional status of the patient with severe paravalvular regurgitation, who was in New York Heart Association (NYHA) class II at 1-year follow-up .
THV implantation should, however, be considered with caution in these particular rigid and open rings. In this setting, semirigid or flexible rings are more favourable as they are easier to circularize during prosthesis deployment, which limits paravalvular regurgitations. However, they are used less often in the tricuspid position.
These emerging techniques of valve-in-valve and valve-in-ring interventions are currently limited to extremely high-risk patients; until now, 150 patients have been treated around the world. These new percutaneous treatments have been proposed for two very different populations: young patients with congenital heart diseases who have already been operated on several times, or older patients with co-morbidities and acquired valve disease who are exposed to the risk of a redo surgery at least. In all cases, patients either had a contraindication to surgery or were considered at very high-risk after evaluation by the heart team.
Given the limited experience with THV implantation in the tricuspid position, the patients who benefit from these techniques should have a close follow-up, including at least an annual visit with TEE and CT scan evaluation in the centre where they were implanted. It is mandatory to ensure that the implanted prosthesis remains functional over a number of years, without any deinsertion, deterioration or thrombosis, before considering larger indications in the future.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
