• Flexibility of the TA and the surrounding myocardium (calcification is rare) counteracts fixation and long-term stability of transcatheter tricuspid valve replacement devices and increases the risk of dehiscence of annuloplasty devices.
• The TA is significantly larger than all other valves with an average area of 21 cm2 in patients with RV dilation.
• The valve is adjacent to important structures including the ostium of the coronary sinus, the AVN and His bundle, the vena cavae, and the right coronary artery (Fig. 1). In surgical repair, direct visualization of the valve apparatus facilitates avoidance of AVN injury (by appropriate positioning of e.g. Kay bicuspidization sutures, De Vega suture annuloplasty and open annuloplasty bands). Such direct visualization is lacking in transcatheter interventions which require a multimodality imaging guidance to minimize the risk of injury of adjacent structures.
• The loss of anatomical landmarks under pathologic conditions (right atrial and ventricular dilation) complicates catheter navigation and interferes with proper positioning of repair/replacement devices.
• The angulation between the annular plane and the superior and inferior venae cavae complicates the transvenous access.
• The RV is thin-walled and the risk of device entanglement in the subvalvular region is high due to the presence of complex subvalvular chordal attachments and RV trabeculations and muscle bands. These factors preclude transapical access.
• Pacemaker/defibrillator leads might interfere with device delivery, deployment, and/or function.
• The low pressure and slow flow in the right heart chambers might provoke device thrombosis.
Important structures adjacent to the tricuspid valve apparatus (as seen in a surgical view through a standard right atriotomy). Reproduced with permission from Shinn et al. . A anterior, Ao aorta, AVN atrioventricular node, CS coronary sinus, FO foramen ovale, IVC inferior vena cava, MS membranous septum, P posterior, RAA right atrial appendage, RV right ventricle, S septal, SVC superior vena cava. Figure 2. Approaches for transcatheter tricuspid valve interventions
Approaches for Transcatheter Tricuspid Valve Interventions
Although relatively recently introduced, many approaches are now available for TTVI and are under preclinical/clinical evaluation. Table 18.2 and Figure 18.2 summarize these approaches .
Approaches for of transcatheter tricuspid valve interventions
a) In native TV (dedicated and non-dedicated devices)
b) In surgical bioprosthesis or in surgical annuloplasty ring
2) Heterotopic (vena caval):
a) Dedicated self-expanding caval valves
b) Off-label balloon-expandable transcatheter heart valves
1) Annular reduction/remodeling devices:
a) The Mitralign device
b) The TriCinch system
c) The Millipede annular ring
d) The transatrial intrapericardial tricuspid annuloplasty (TRAIPTA) device
e) The Cardiac Implants annuloplasty device
2) Coaptation devices:
a) Regurgitant orifice occupation (the FORMA Repair System and the TV Occluder device)
b) Edge-to-edge repair (the MitraClip device)
Examples of transcatheter repair and, heterotopic, valve implantation approaches to treat tricuspid regurgitation. In the lower panel, different examples of transcatheter repair devices are displayed. Reproduced with permission from Rodés-Cabau et al.  and Latib et al. J Am Coll Cardiol. 2017;11;69:1807–1810.
Orthotopic Valve Implantation
Native Tricuspid Valve Replacement
Dedicated Transcatheter TV Bioprosthesis
Boudjemline et al. developed a transcatheter bioprosthetic valve dedicated for fixation to the native TA . The self-expandable double-disc nitinol stent consisted of two flat disks (spontaneous diameter = 40 mm) and a tubular portion (spontaneous diameter = 18 mm), with an overall length of 15 mm when deployed. The whole device is braided from a single nitinol wire (i.e. all parts are connected in one unit). Animal implantations were performed through the right jugular vein and deployment is similar to devices for closure of atrial septal defects (RV disk deployment and traction against the TA, followed by deployment of the leaflet-containing tubular part, and finally the atrial disk). In eight ewes (maximum TA diameter: average, 30 mm; range, 27–35 mm), deployment was attempted and failed in one (entrapped in the subvalvular apparatus leading to underexpansion) while one prosthesis showed a significant paravalvular leakage . The device was not clinically used.
The GATE™ tricuspid atrioventricular valved stent (NaviGate Cardiac Structures Inc., CA) is another dedicated transcatheter TV bioprosthesis (available in sizes: 30/40, 33/44 and 36/48 mm) which was successfully implanted in an animal model through a beating-heart minimally-invasive surgical and transjugular percutaneous implantation . The first in man implant took place in November 2016 .
Non-Dedicated Transcatheter Heart Valve Implantation in the Tricuspid Position
Very few clinical attempts were made to implant a transcatheter heart valve (THV) in the native TA. Kefer et al. reported balloon-expandable aortic THV implantation to replace a native TV in the absence of prosthetic material and radiographic landmarks . Implantation was performed under general anesthesia and extracorporeal membrane oxygenation, through a femoral vein access. Balloon sizing was used to decide the THV size and the TA was pre-stented with two covered stents. Implantation of a 26-mm Sapien valve (Edwards Lifesciences, Irvine, CA) was followed by another 26-mm Sapien (valve-in-valve) implantation due to paravalvular leakage of the first. The clinical and echocardiographic outcomes were favorable up to 5 months.
Transcatheter Valve in Malfunctioning Surgical Bioprosthetic Valve
There are many reports on successful implantation of the self-expanding Melody valve (Medtronic Inc., Minneapolis, Minnesota) and the balloon-expandable Sapien, Sapien XT, and Sapien 3 valves (Edwards Lifesciences, Irvine, CA) in malfunctioning stented [53–55] and stent-less  bioprosthetic tricuspid valves. Given the small sizes available for the Melody valve (≤22 mm), it has been utilized mainly in children or young adults with CHD . The first implants were done mainly via a transatrial or transjugular access. More recently, the transfemoral access under local anesthesia and fluoroscopic guidance or general anesthesia and transesophageal echocardiography (TEE) guidance was shown to be feasible and safe. Unless the patient has a previously implanted permanent pacemaker, rapid pacing (which is usually needed for valve deployment) is performed by positioning the temporary lead in the left ventricle or in the coronary sinus. Predilatation of the bioprosthesis is not recommended to avoid the risk of leaflet rupture and embolization . Deciding the size of the transcatheter valve to be implanted within the degenerated bioprosthesis, or within the dysfunctioning annuloplasty ring, should be guided by the nominal internal degenerated device area, by multi-modality imaging planning (combining computed tomography and echocardiography), and by balloon-sizing. Due to lower closing pressures, oversizing of the prosthesis is considered to be less crucial in the tricuspid position (as compared to aortic and mitral positions) .
A large recently published multicenter series included 152 patients who underwent implantation of a THV within a malfunctioning surgical bioprosthetic valve . Melody valve was used in 94 and Edwards valves in 58 (Sapien in 12, Sapien XT in 41, and Sapien 3 in 5). Procedural success was achieved in 98% of attempts, and freedom from residual TV dysfunction (significant stenosis and/or regurgitation) was achieved in 97%. Five patients (3%) died and one patient underwent TV reintervention (due to valve thrombosis) within 30 days of implantation. During follow up (median: 13.3 months), 17 additional deaths (11%) and 10 TV re-interventions (7%; 2 due to TR, 2 due to TV stenosis, 2 due to combined TR and stenosis, 2 due to multiorgan failure, 1 due to endocarditis, and 1due to cardiac transplant graft failure) took place. Four patients (all with a Melody valve) were diagnosed with endocarditis and four patients (3 Melody, 1 Sapien) were diagnosed with sterile thrombus or immobility or thickening of the valve.
Transcatheter Valve in Tricuspid Repair Ring
Although fewer cases have been reported than valve-in-valve implantations, transcatheter implantation of tricuspid valve-in-ring was shown to be feasible and safe. There have been reports of implantation of Sapien XT and Melody  THVs in Carpentier-Edwards classic and rigid  annuloplasty rings through transfemoral [58–60] and transatrial (off-pump through a minithoracotomy)  approaches. Some relevant technical considerations are worth mentioning. The ring should be used as a fluoroscopic landmark and paravalvular leak (due to the non-circular landing zone created by the ring, especially the incomplete rings) should always be anticipated and effectively managed .
Heterotopic (Vena Caval) Valve Implantation
Implantation of percutaneous valves into the caval veins targets amelioration of the systemic venous congestion without correcting TR itself. Therefore, the risk of progressive RV and right atrial (RA) deterioration, RV failure, and AF persists. Although intracardiac manipulations are not needed and the valve is implanted in a vascular tube rather than in a complex valvular apparatus, fixation of a rigid percutaneous valve within a highly compliant thin-walled vascular tube is challenging. Other challenges include, the large variable diameters of inferior and superior vena cavae (IVC and SVC ) in the presence of chronic severe TR and the proximity to the RA and to hepatic veins. Demonstration of pulsatile blood flow and systolic flow reversal in the caval veins are prerequisites for the proper function of the caval valves. Because of existing pacemaker/implantable cardioverter-defibrillator leads in many cases and the lack of evidence of clinically-relevant SVC congestion in most cases, 90% of implants are done in the IVC only. Two currently available technologies involve vena caval valve implantation:
Self-Expanding (Dedicated) Valve
The TricValve (P&F Products & Features Vertriebs GmbH, Vienna, Austria, in cooperation with Braile Biomedica, São José do Rio Preto, Brazil) consists of two separate tri-leaflet pericardial tissue valves which are mounted on a self-expanding nitinol stent . The size ranges from 28 to 43 mm (IVC valve) or 38 mm (SVC valve). The upper segment of the IVC valve protrudes into the RA while the lower segment is located above the diaphragm to avoid occlusion of the hepatic veins. The SVC valve stent has a funnel-shaped design to fit to the superior cavo-atrial junction. Both IVC and SVC valves are delivered transfemorally through a dedicated flexible catheter. So far, clinical implantations are limited to compassionate use in a few patients [63, 64]. In a patient who was followed up for 12 months after implantation, the functional status was significantly improved and the clinical signs of right-side HF were completely resolved, while echocardiography revealed excellent valve function, unchanged position, and no paravalvular leakage .
Balloon-Expandable Transcatheter Heart Valves
Off-label clinical use of Edwards Sapien, Sapien XT or Sapien 3 valves implanted into the cavo-atrial junction has been reported [65, 66]. The large diameter of the caval veins in the presence of severe chronic TR, the lack of calcification, and the confluence of hepatic veins preclude direct implantation and require pre-stenting of the landing zone. In cases with markedly dilated caval veins, minimally-invasive surgical banding can help reduce the lumen diameter by wrapping the vein with a longitudinally opened Gore-Tex prosthesis simultaneously with balloon inflation within the vein to control lumen reduction . In a small case series, valve function remained excellent with no transvalvular or paravalvular regurgitation up to 30 days after implantation. All patients experienced an improvement of their dyspnea and of the clinical signs of systemic venous congestion, as well as improvement of RV function, RV and RA volumes, and the diameters of the hepatic veins .
Transcatheter Tricuspid Valve Repair
Annular Reduction/Remodeling Devices
Understanding the mechanism of incompetence in functional TR (basically, annular dilation and remodeling) and the favorable outcome of annuloplasty compared to other surgical repair techniques of TR , jointly led to a significant interest in developing transcatheter annuloplasty techniques developing transcatheter annuloplasty techniques to traet TR. These techniques generally reproduce the established surgical annuloplasty approaches, e.g. Kay bicuspidization (Mitralign) and ring annuloplasty (Millipede). The ideal annuloplasty device to treat high-grade functional TR should restore the normal three dimensional elliptical shape of the TA to reduce leaflet stress and tethering. It should tackle the selective propensity of the anterior-posterior- commissure sector of the annulus to remodel (along the RV free wall) and also be ‘open’ at its septal- leaflet sector to spare the atrio-ventricular conduction system . Ideally, it should also be flexible maintaining annular dynamicity and preventing ring dehiscence .
The Mitralign/Trialign Device
The Mitralign device (Mitralign, Tewksbury, Massachusetts) is a percutaneous annuloplasty system that reproduces the Kay bicuspidization surgical procedure , which converts an incompetent tricuspid into a competent bicuspid valve by plicating the posterior leaflet. Mitralign system comprises an articulating wire delivery catheter, a pledget catheter (preloaded with 4 × 8 mm polyester pledgets), and a stainless steel plication lock. Transjugular access with two sheaths is obtained, and the steerable catheter is advanced across the TV into the RV and is articulated under the annulus to position its tip beneath either the anteroposterior or the septoposterior commissure. An insulated radiofrequency wire is advanced to burn through the annulus (2–5 mm from the hinge point of the leaflet) into the RA and is then externalized and replaced by the pledget delivery catheter. Two pledgets (2.4–2.8 cm apart) are usually released and are cinched onto the annulus and finally are brought together using the plication lock (Fig. 18.3). Right coronary artery wiring is performed at the start of the procedure and angiography is performed at its end . In a number of published clinical implantations, the procedure resulted in a significant reduction in annular circumference (24%)  and area (57%)  and effective regurgitant orifice area (53%) . Although immediate complete elimination of TR was reported in some cases , acute recurrence due to pledget dehiscence has also been observed in some cases . The importance of the risk of pledget dehiscence and TR recurrence was further confirmed by the early results of the Percutaneous Tricuspid Valve Annuloplasty System [PTVAS] for Symptomatic Chronic Functional
Summary of the procedural steps of the Trialign transcatheter annuloplasty technique: Through a deflectable guide catheter, a wire delivery catheter followed by an insulated radiofrequency wire (a) and pledgeted sutures (c) are advanced through the tricuspid annulus. The 2 sutures are plicated and locked (e) and then cut leaving a bicuspidized valve (g). b, d, f, and h are 3D transesophageal echocardiographic images corresponding to the a, c, e, and g cartoons. Reproduced with permission from Hahn et al. . Ant anterior, Post posterior, RA right atrium, RV right ventricle, Sept septum
Tricuspid Regurgitation (SCOUT ) study, where three patients out of 15 (20%) had a single-pledget annular detachment within 30 days after repair . The study included 13 females and two males, with an average age of 74 years who had moderate-severe functional TR (average vena contracta diameter, 13 mm). Patients with severe pulmonary hypertension or severe right or left ventricular systolic dysfunction were excluded. Two thirds of patients had prior mitral valve intervention and two thirds had AF. Immediate procedural success was achieved in all 15 cases, with no procedural death or major complication with the exception of one case of right coronary artery compromise (treated with a stent placement). At 30 days, three patients had an echocardiographic evidence of single pledget detachment and showed no improvement of TR severity, but required no reintervention. In the remaining 12 patients, tricuspid annulus area (12.3 ± 3.1 cm2 vs. 11.3 ± 2.7 cm2, p = 0.019) and TR vena contracta diameter (13 ± 4 vs. 10 ± 3 mm, p = 0.022) were modestly reduced while the quality of life was significantly improved at 1 month after repair.
Although technically challenging with risk of dehiscence especially with inadequate size or unfavorable tissue quality of the posterior annular shelf, this device has a small footprint and largely preserves the underlying anatomy and the procedure can be customized/repeated to address different severities of annular dilatation. Moreover, a controlled ‘partial’ correction of TR is possible, thus avoiding the acute hemodynamic burden on the RV of a sudden complete correction of a severe long-standing TR.
The TriCinch System
The TriCinch system (4Tech Cardio, Galway, Ireland) consists of a Dacron band connecting a corkscrew anchor attached anterior to the TV annulus (typically between the anteroposterior commissure and the mid-point of the anterior leaflet) to a self-expanding subhepatic IVC stent. By pulling the anchor towards the IVC, by means of the Dacron band, the TA antero-posterior and antero-septal dimensions are remodeled, and the tension is maintained by fixation of the stent in the IVC. The stent is 60 mm long and is available in a range of diameters (from 27 to 43 mm) to allow for adequate oversizing in the IVC . The device is retrievable and the procedure is relatively easy to perform. Right coronary artery wiring and ad-hoc angiography are usually needed to exclude coronary injury. Hemopericardium, failure of implantation, and anchoring system detachment lead to failure of the procedure in up to 50% of attempts .
This approach of TA remodeling, changes the geometry rather than reduces the size of the annulus, leading to an unpredictable extent of TR reduction. In one case report, TA septolateral diameter decreased from 41 mm to 38 mm, and TR severity improved from 4+ to 3+ . In another case report, severe TR was reduced to mild, with a 50% reduction of the effective regurgitant orifice area . In a third case , severe TR was reduced to mild, and the procedure was performed under sedative anesthesia and intracardiac and transthoracic (but not transesophageal) echocardiographic guidance.
The Transatrial Intrapericardial Tricuspid Annuloplasty (TRAIPTA) Device
TRAIPTA is an external (epicardial) nitinol annuloplasty loop which is wrapped around the atrioventricular groove. The pericardial space is reached via a puncture in the RA appendage, which is closed with a percutaneous occluder device at the end of the procedure. The loop is passed around the cardiac apex and then retracted until it encircles the atrioventricular groove, and is eventually tightened using a wire within it . The TRAIPTA device has not been used in humans. Animal studies were performed under general anesthesia and mechanical ventilation and the suture was tightened under guidance by real-time magnetic resonance imaging and intracardiac echocardiography .
Postimplantation magnetic resonance imaging demonstrated a 49% reduction of the septolateral TA diameter, a 31% reduction in the anteroposterior diameter, and a 59% reduction of TA area. The effect on the mitral valve annulus was modest (15% diameter reduction), and a non-hemodynamically significant pericardial effusion that resolved spontaneously was seen after all swine implantations .
Reproducing these results in humans is, however, doubtful. The spatial orientation of the tricuspid and mitral annuli relative to the atrioventricular groove is different in humans than in swine and is variable from patient to patient. In most human cases, at least one epicardial coronary artery crosses the projected course of the TRAIPTA implant . The high RA pressure in patients with long standing TR may increase the risk of tamponade, and pericardial adhesions from previous cardiac surgery may preclude safe epicardial access in many patients .
Surgical-Like Annuloplasty Rings
This is a group of transcatheter annuloplasty rings that aim at reproducing the surgical ring annuloplasty approach. In addition to the proven superiority of ring annuloplasty over other approaches of surgical TV repair , this approach has the potential of combination with a coaptation repair procedure (in the same session or ad-hoc) and allows for and facilitates subsequent valve-in-ring implantation; by acting as a landmark and by securing transcatheter valve fixation. Flexible and adjustable complete (Millipede™  and Cardiac Implants™ ) and incomplete (CardioBand™ ) annular rings are at different stages of clinical testing.
The Cardiac Implants annuloplasty device is implanted on the atrial surface of the tricuspid annulus and is awaited to heal for 6–8 weeks by tissue reaction. After that, the device can be cinched to reduce the TA diameter or be used as a platform for valve-in-ring implantation .
In addition to annular dilatation, leaflet tethering and malcoaptation are the second most important mechanisms of functional TR. Two basic concepts make the platform for the coaptation devices, occupation of the regurgitant orifice and leaflet edge-to-edge clipping.
The FORMA Repair System and the TV Occluder Device
The transcatheter Forma Repair System (Edwards Lifesciences, Irvine, California) is designed to reduce TR by occupying the regurgitant orifice and providing a surface for native leaflets coaptation. It is composed of a rail, which is anchored at the apex of the RV, and a spacer, which serves as the coaptation element. The spacer consists of a foam-filled polymer balloon and is currently available in two sizes (12 and 15 mm), with a length of 42 mm. The spacer expands passively within the vascular system to its final size by air through eight vents in its shaft. Access is via the left subclavian or axillary vein, which should be sizable to accommodate a large introducer. Following achievement of a satisfactory degree of TR reduction, the device is locked within the subclavian region, and extra rail length is placed into a subcutaneous pocket. The entire device is fully retrievable during all stages of the procedure. Seven compassionate clinical implants have been published, all the seven had residual moderate TR at 30 days post-repair . It should be noted however that the echocardiographic assessment of residual TR after spacer implantation between the leaflets is challenging, given the resulting distortion of the regurgitant orifice and the vena contracta.
The TV Occluder Device (Cleveland Clinic Foundation, Cleveland, Ohio) is utilizing the same concept on which the Forma system is based; occupation (and consequently reduction) of the regurgitant orifice area and facilitating leaflet coaptation . Similar to the Forma device, the TV Occluder system has a proximal end that is left in a subcutaneous pocket (similar to a pacemaker). This allows for an extended potential for control of the degree of the regurgitant orifice occlusion. The distal end of the system consists of a screw that is parked in the RV apex. The occluder itself is made of a nitinol mesh skeleton and an internal membrane of polyester fabric. Small-scale animal studies (through a transjugular approach) show a reasonable degree of TR reduction .
The MitraClip Device
Surgical edge-to-edge tricuspid valve plasty was shown to be an effective adjuvant procedure for patients who have severe residual TR after traditional ring repair . The transcatheter MitraClip system (Abbott Vascular, Santa Clara, CA, USA) has been established as an effective treatment of mitral regurgitation in high risk patients . The device is a 4-mm-wide cobalt-chromium, polyester-covered clip that can be opened and closed by control mechanisms on the clip delivery system. Multiple reports of TV transcatheter clipping using the MitraClip system are published [84–88].
In an ex vivo model of functional TR in a porcine heart connected to a pulsatile pump, the effectiveness of the edge-to-edge technique using the MitraClip device was tested . Transvenous valve clipping led to an increase in the RV stroke volume without a significant increase in the transvalvular gradient. Procedural success was strongly influenced by the position where the clips were applied (medial vs. commissural) and the pair of leaflets being grasped. Clips that involved the septal leaflet in the medial position were the most effective in restoring the physiological hemodynamics (cardiac output, mean pulmonary pressure, and mean diastolic valve pressure gradient), while those that did not involve the septal leaflet (i.e. grasping the anterior and posterior leaflets) did not reduce TR  (Fig. 18.4).
Evaluation of the effectiveness of transcatheter edge-to-edge treatment of functional tricuspid regurgitation in an ex vivo model: (Left) In functional tricuspid regurgitation, the tricuspid valve annulus expands more towards the right ventricular free wall (i.e. the anterior (A) and posterior (P) leaflets are pulled (arrows) away from the septal (S) leaflet. (Right upper) In an ex vivo porcine model of tricuspid regurgitation, the clip is applied to the S and A leaflets (in a medial position) restoring the coaptation lines between all the 3 leaflet-pairs, with consequent restoration of valve competence. (Right lower) The clip is applied to the A and P leaflets (in a medial position) leaving the gap between A–P and S leaflets unaltered and the regurgitation persistent. (Reproduced and modified from Vismara et al. )
In clinical human cases [84–88], femoral and jugular venous approaches have been used, and clips were deployed at the anteroposterior, anteroseptal, and/or posteroseptal lines of coaptation. Although the technique is simpler than most other TTVIs, two challenges of TV clipping should be considered. First, the acute angle between the IVC and the TA plane makes coaxial positioning difficult. This can be overcome by using the transjugular approach, which is, however, unsuitable for the large sheath in some patients. Secondly, regurgitation can develop at any point along the line of coaptation between the three leaflets of the TV valve. Clipping all three leaflets might be required and increases the risk of inducing valve stenosis. Moreover, more than three leaflets can be identified in many patients.
Open Issues and Future Perspective
While TTVIs are generally promising, there remain several important issues to be addressed:
The risk of device thrombosis : Devices in the TV position may be at a significant risk of thrombosis due to the lower intracardiac pressures in the right side of the heart . However, as per practice guidelines, long term anticoagulation is not recommended in patients with surgical bioprosthetic valves, even those present in the right side of the heart. In the published series of three patients treated with the caval valve implant (Sapien XT) , routine anticoagulation was employed and no cases of thrombosis were observed. In the tricuspid valve-in-valve implantation series by McElhinney et al. (n = 152), the proportion of the overall cohort treated with anticoagulation is not reported; however four patients (2.6%) developed severe valve obstruction due to valve thrombosis, 2 while on aspirin, 1 on warfarin, and 1 on both aspirin and warfarin . In patients treated with non-valve replacing TV repair devices such as the Forma, Mitralign, and TriCinch devices, the need for anticoagulation is even less clear. However, all but one patient received anticoagulation in the Forma early feasibility study , possibly due to the high prevalence of AF requiring anticoagulation. Notably, in the feasibility study of the Mitralign device (SCOUT), two thirds of the patients had AF . Although these observations cannot lead to any conclusions, they obviate the need for longitudinal studies that focus on this question. It is important to note that, unlike left-side valve devices where thrombosis typically presents with sudden systemic embolism, right-side valve device thrombosis tends to present with obstructive manifestations. Therefore, a rise in transvalvular velocity and gradient on serial Doppler surveillance should serve as a surrogate marker of device thrombosis. Although echo-Doppler can suspect/detect thrombus formation on a bioprosthetic valve or a coaptation device, this can be more challenging in annuloplasty devices. Whether computed tomographic (CT) scanning can fill the gaps in this surveillance process is not yet investigated.