Transcatheter pulmonary valve implantation

INTRODUCTION — Transcatheter pulmonary valve implantation (TPVI) is an alternative to surgery for selected patients with right ventricular outflow tract (RVOT) dysfunction [1-3]. TPVI devices were initially developed to treat dysfunctional (stenotic or regurgitant) RV to pulmonary artery conduits. Indications were later expanded to dilated RVOTs following patch placement as valves were introduced for this application. TPVI is intended to reduce the total number of open-heart surgeries required over a patient's lifetime.

CLINICAL SETTING

Uses — TPVI is a treatment for selected patients with RV outflow tract (RVOT) dysfunction involving a bioprosthetic or native pulmonary valve; the valve may be located in an RV to pulmonary artery conduit or in a native RVOT [2-8]:

●Pulmonary regurgitation with associated criteria and mild or no nonvalvular RVOT obstruction, as discussed separately (table 1). (See "Tetralogy of Fallot (TOF): Long-term complications and follow-up after repair", section on 'Indications' and "Pulmonic regurgitation", section on 'Intervention'.)

●Pulmonary valve stenosis with associated criteria for intervention, which are discussed separately. (See "Tetralogy of Fallot (TOF): Long-term complications and follow-up after repair", section on 'Indications' and "Pulmonic valve stenosis in adults: Management", section on 'Indications for intervention'.)

Underlying conditions — TPVI is performed in selected patients with RVOT/pulmonary valve dysfunction in the following clinical settings:

●Repaired native RVOT – Conditions that may be treated with RVOT repair include tetralogy of Fallot (TOF) and pulmonary atresia. (See "Tetralogy of Fallot (TOF): Management and outcome" and "Tetralogy of Fallot (TOF): Long-term complications and follow-up after repair".)

●RV to pulmonary artery conduit – Conditions that may be treated with an RV to pulmonary artery conduit (valved or nonvalved) include TOF, pulmonary atresia, and truncus arteriosus. (See "Tetralogy of Fallot (TOF): Management and outcome" and "Tetralogy of Fallot (TOF): Long-term complications and follow-up after repair" and "Pulmonary atresia with intact ventricular septum (PA/IVS)" and "Truncus arteriosus".)

●Bioprosthetic pulmonary valve dysfunction – TPVI can also be performed as a valve-in-valve procedure after failure of the primary bioprosthetic valve in the RVOT [9]. The initial bioprosthetic valve may have been placed for RVOT dysfunction or as part of a Ross procedure. The indications and the implantation techniques are the same for repeat TPVI as for primary TPVI.

●Native pulmonary valve stenosis or regurgitation – For patients with native pulmonary valve disease who require valve replacement, TPVI is an alternative to surgical valve replacement. (See "Pulmonic valve stenosis in adults: Management", section on 'Interventions' and "Pulmonic regurgitation" and "Pulmonic regurgitation", section on 'Percutaneous'.)

●Failed pulmonary balloon valvuloplasty or surgical pulmonary valvotomy – TPVI is a potential option for selected patients with significant residual pulmonary valve stenosis or pulmonary regurgitation after balloon valvuloplasty or surgical valvotomy for pulmonary valve stenosis.

Choice of procedure — Selection of procedure for RVOT dysfunction (pulmonary regurgitation and pulmonary valve stenosis) is discussed in separate topic reviews. (See "Tetralogy of Fallot (TOF): Long-term complications and follow-up after repair", section on 'Pulmonary valve replacement (PVR)' and "Pulmonic regurgitation" and "Pulmonic valve stenosis in adults: Management".)

When intervention is indicated for RVOT dysfunction, and TPVI is feasible, TPVI is commonly preferred to surgical pulmonary valve implantation to minimize procedural risk and recovery time. However, when there is a concurrent indication for a cardiac surgery (such as tricuspid annuloplasty, coronary artery bypass, or arrhythmia surgery) surgical pulmonary valve implantation is generally preferred.

Exclusions — TPVI is not performed in the following settings due to technical and/or safety concerns, as discussed in the 2020 European Society of Cardiology guidelines for the management of adult congenital heart disease endorsed by the Association for European Pediatric Cardiology [2]:

●Severe pulmonary valve stenosis that cannot be relieved by balloon dilation (eg, severely stenotic dysplastic valve (thickened cusps and redundant tissue as seen in Noonan syndrome).

●RVOT (native or conduit) size that exceeds the limit of available devices. The RVOT size must be small enough to provide a suitable landing zone for a covered stent. There is generally no lower limit for landing zone size.

•For balloon-dilatable percutaneous valves, a landing zone diameter <29 mm is generally required. In selected cases, a slightly larger diameter may be successfully treated after prestenting to a smaller diameter to create a suitable "landing zone" or by overdilating an Edwards SAPIEN valve 29 to 31 mm.

•Self-expanding pulmonary valves enable treatment of larger RVOT anatomies. (See 'Self-expanding valves' below.)

●Inadequate venous access from both femoral and jugular sites caused by:

•Conditions such as central vein occlusion.

or

•Small body size (eg, body weight ≤20 kg). Femoral vein (or jugular vein size must be adequate for placement of the introducer.

●Risk of coronary compression by the expanded implant as determined by balloon testing (image 1) [10]. (See 'Balloon test' below.)

●Active systemic infection (such as endocarditis) or high risk of systemic infection (such as intravenous drug abuse).

PREPROCEDURAL EVALUATION — A multidisciplinary assessment is performed to determine whether the patient is a candidate for pulmonary valve replacement, choice of procedure (TPVI versus surgery) and choice of valve.

●Criteria for pulmonary valve replacement (for pulmonary regurgitation and/or pulmonary valve stenosis) are assessed by clinical evaluation, echocardiography, cardiovascular magnetic resonance imaging (CMR), and/or cardiac computed tomography (CT), as discussed separately. (See "Tetralogy of Fallot (TOF): Long-term complications and follow-up after repair", section on 'Indications'.)

●Additional role of imaging for TPVI – For patients undergoing TPVI, coronary CT angiography, CMR, and/or invasive coronary angiography may be used to assess the spatial relationship between the coronary arteries and the RVOT [2], although this is further assessed by the balloon test. (See 'Balloon test' below.)

For self-expanding valves in native RVOTs, high-quality preinterventional three-dimensional imaging (by cardiac CT or CMR) is performed in systole and diastole. These images inform selection of a balloon size for the balloon test. (See 'Balloon test' below.)

TPV DEVICES — For patients undergoing TPVI, the choice of transcatheter pulmonary valve (TPV) is based upon availability and RV outflow tract (RVOT) size (with self-expanding valves used for the largest RVOTs, such as native RVOTs).

Balloon-expandable valves — Two balloon-expandable TPVs are US Food and Drug Administration (FDA) and CE approved. Depending on the final diameter of the treated RVOT, either a Melody valve or a SAPIEN 3 valve is chosen.

●Melody TPV – This valve (image 2) is made from a Contegra bovine jugular vein valve hand-sewn into a 34 mm bare metal Cheatham platinum balloon-expandable intravascular stent (NuMED CP Stent CP8Z34) and is available in 2 sizes: a 16 mm valve that can be expanded up to 20 mm and an 18 mm valve that can be expanded up to 22 mm [11].

●SAPIEN 3 TPV – The valve (image 3) is made from a trileaflet bovine pericardial valve hand-sewn into a balloon-expandable stainless steel stent and is available in three sizes, 23, 26, and 29 mm, and is approved for RVOT conduit diameters from 16.5 to 29 mm (16.5 to 28.5 mm internal diameter of a failed bioprosthesis) [12]. There is a fabric-sealing cuff covering the proximal part of the stent designed to prevent paravalvular leaking.

Self-expanding valves — Two balloon expandable valves are US FDA and CE approved:

●Harmony TPV – This is a self-expanding porcine pericardial valve mounted within a pericardial tube that is supported by a nitinol framework. The valve is specifically designed for use in large-diameter native outflow tracts.

The Harmony TPV is available in two designs. It consists of a self-expanding nitinol stent with a woven polyester covering and a porcine pericardial trileaflet valve sewn into its center. The various inflow and outflow sizes are designed to accommodate a variety of native outflow tract morphologies [13]. The valve housing is 22 mm (inflow diameter 41 mm; exit diameter 32 mm; total length 55 mm) or 25 mm (inflow diameter 54 mm; exit diameter 43 mm; length 51 mm).

●SAPIEN 3 with Alterra adaptive prestent – This is a two-part valve system designed for valve implantation in large-diameter native RV outflow tracts. The Alterra adaptive prestent is an anchoring adaptor for the 29 mm SAPIEN 3 valve within native RVOTs.

The prestent is made of a self-expanding, radiopaque, nitinol frame assembly covered proximally by polytetrafluoroethylene. The prestent flares symmetrically at inflow and outflow ends (40 mm diameters). The central waist of the prestent was designed to provide a stable landing zone in a dilated RVOT for a 29 mm SAPIEN 3 transcatheter valve, which is typically implanted during the same procedure [14]. The total length of the Alterra prestent is 48 mm.

Other self-expanding valves

●Venus P-Valve – This valve is composed of a self-expandable nitinol framework with a trileaflet porcine pericardial valve and is available in straight and flared designs in five sizes (central stent diameter 28 to 36 mm in 2 mm increments) and two lengths (central stent length 25 and 30 mm). It is loaded on a 22 or 24F delivery system. The valve is preferably implanted through a 26F 65 cm DrySeal sheath. The valve underwent several modifications since its first implantation. In 2023 it received CE mark for Europe. This valve is an investigational device in the United States.

●Pulsta TPV – This trileaflet valve consists of a double-strand nitinol wire covered with treated porcine pericardium. Valve central diameters range from 18 to 32 mm, again with 2 mm increments (thus including the smallest diameter for a self-expanding valve). Both valve end diameters flare by 4 mm greater than the central diameter. The total valve length is 28 to 38 mm. The valve is approved in South Korea and is awaiting CE certification in Europe.

IMPLANTATION TECHNIQUE — TPVI is usually performed using general endotracheal anesthesia or conscious sedation. Venous access is generally obtained through the femoral vein, although a jugular venous approach is also feasible. After a complete hemodynamic and angiographic study, a guidewire is positioned distally into a pulmonary artery.

Balloon test — Since coronary artery compression is a life-threatening complication of TPVI, a balloon test is performed in the RV outflow tract (RVOT) to exclude risk of this complication. This involves inflation of a balloon in the RVOT (conduit or native). The balloon test is performed prior to TPVI with any balloon expandable valve. Although coronary compression does not seem to occur often with self-expanding transcatheter pulmonary valves (TPV), a balloon test with aortogram is also frequently performed prior to TPVI with self-expanding valves.

●The balloon size is selected based upon the RVOT landing zone for the pulmonic valve. (See 'Preprocedural evaluation' above.)

In enlarged "native" RVOTs, an RV angiogram is obtained to document total occlusion of the RVOT during balloon inflation. Based on the balloon test, an adequately sized valve is chosen, usually 2 to 4 mm larger than the waist seen in the balloon during interrogation of the landing zone.

●The balloon test is initially performed with a low-pressure balloon to reduce the risk of conduit rupture. A balloon is inflated with a simultaneous aortogram to display the coronary arteries (image 1) [10]. For this test, an aortogram may be preferable to a selective coronary artery angiogram because engaging the coronary artery with the catheter may lead to a false negative test.

If the origins of the coronary arteries are in close proximity to the RVOT, a balloon test with an ultra-high-pressure balloon (ie, Atlas Kevlar) may be necessary to avoid coronary artery compression and selective coronary angiography is performed in multiple planes.

If coronary artery flow is impaired during the balloon test, TPVI should not be performed.

Conduit preparation and prestenting — If there is significant conduit stenosis, conduit balloon angioplasty and stenting is performed to reduce the RVOT gradient.

The role of prestenting varies depending upon the type of valve:

●Melody valve – Prestenting of the conduit before Melody valve implantation is standard practice at most centers since observational studies have found that prestenting is associated with a lower risk of TPV stent fractures [15]. Usually, a covered stent is preferred to avoid bleeding related to conduit rupture during stent deployment because high-pressure balloons are needed to dilate the conduit to the previously tested diameter. In some cases (eg, long stenosis or significant recoil during balloon deflation), additional stents are needed for prestenting to support the landing zone. A noncovered stent is usually used if obstruction of the origin of one of the pulmonary branch arteries ("jailing") is expected after deployment of the stent.

●SAPIEN valve – In selected patients, the SAPIEN valve may be implanted directly without prestenting since stent fracture has not been reported [16].

●Self-expanding valves - Prestenting is not necessary for self-expanding valves (such as Harmony and Venus P-Valve) and may even be counterproductive.

Valve implantation — The implantation techniques for the Melody and SAPIEN valves are similar [1,7,17-20].

●The Melody valve is crimped onto a designated double balloon delivery system requiring a 22F sheath.

●The SAPIEN valve is crimped onto a designated balloon delivery system requiring a 14 or 16F sheath for the 23, 26, or 29 mm valve. The SAPIEN THV is advanced without protection into the RVOT. If the THV struts get caught in the tricuspid valve, the valve may be damaged [21]. To avoid this complication, a large sheath is used for direct implantation of the valve in the RVOT [22].

●Self-expanding valves are crimped and loaded into a delivery catheter system. The assembly is positioned over a stiff wire and then the delivery catheter is slowly withdrawn as the self-expanding valve opens. The correct valve position is depicted via a second catheter in the RVOT prior to final valve delivery.

COMPLICATIONS BY VALVE TYPE — Short- and medium-term results of TPVI with the Melody and SAPIEN valves are similar, although more data are available for the Melody valve. Long-term data are available for the Melody valve, as discussed below. Long-term data are not yet available for the SAPIEN 3 valve and the self-expandable valves.

For balloon-expandable valves — Many studies show that regardless of the type of balloon-expandable valve used for TPVI, higher residual gradient correlates with unfavorable valve outcomes [23-26]. Therefore, every effort should be made to reach the lowest possible RV-pulmonary artery gradient at the end of the procedure.

An international multicenter registry study examined the results for 2476 patients who underwent TPVI with either the Melody or any Sapien valve who were followed for 8475 patient years [26].

●A total of 95 patients died after TPVI, most commonly from heart failure (n = 24) for a cumulative incidence of death of 8.9 percent at eight years after TPVI. On multivariable analysis, age at TPVI, a prosthetic valve in other positions, and an existing transvenous pacemaker or implantable cardioverter-defibrillator were associated with death.

●A total of 258 patients underwent TPV reinterventions. At eight years, the cumulative incidence of any TPV reintervention was 25.1 percent and of surgical TPV reintervention was 14.4 percent. Risk factors for surgical intervention included age, prior infective endocarditis, TPVI into a stented bioprosthetic valve, and postimplant gradient.

Melody TPV — Data are available from four short- and medium-term observational studies with a total of over 450 patients with one- to five-year follow-up [7,8,27,28] as well as from longer-term studies. These studies excluded 4 to 8 percent of the patients who did not undergo TPVI due to an unfavorable RV outflow tract (RVOT) dimension or coronary artery compression.

After TPVI, the following findings were noted:

●Overall operative mortality was 0 to 5 percent and was, as far as could be assessed, not related to the TPVI.

●The device was unstable in 1 percent, a conduit rupture requiring surgery occurred in 0.8 to 2 percent, compression of the left main coronary artery after device deployment occurred in 0 to 1 percent, and obstruction of the origin of one of the pulmonary branch arteries ("jailing") occurred in 0.6 percent.

●The RVOT gradient fell significantly from median values of 35 to 40 to 10 to 20 mmHg. Median peak velocity over the RVOT was 1.9 to 2.7 m/s at one-year echocardiographic follow-up. Pulmonary regurgitation decreased from median values of 16 to 27 percent to 1 to 2 percent.

●Freedom from Melody valve dysfunction or reintervention was 94 to 95 percent at one year. Patients not requiring reintervention had mild or less pulmonic valve regurgitation at one-year follow-up. Overall, surgery was required in 3 percent during follow-up, with freedom from reoperation of 86 percent at 30 months.

●Endocarditis was diagnosed in 1 to 3 percent of the cases up to five years after TPVI with Melody valves. (See 'Endocarditis' below.)

●At six months, New York Heart Association class improved in 77 percent of the patients and worsened only in 0.8 percent. RV end-diastolic volume and RV mass decreased significantly as demonstrated by CMR imaging.

●Stent fractures occurred in 20 to 21 percent in a series without prestenting and in 5 to 16 percent in a series with prestenting. (This contrasts with no reported stent fractures with the Sapien 3 valve.) (See 'SAPIEN TPV' below.)

●In an early study, the first 50 patients of the series and those with residual gradients >25 mmHg had a higher risk of reoperation compared with the rest of the study group [27].

Four long-term studies on TPVI with the Melody valve have been published [23-25,29]. The best long-term results (estimated event-free survival rate of 88 percent at 10 years) were observed among patients with a RVOT residual peak invasive gradient of <15 mmHg after the intervention. One study compared the long-term outcome of TPVI (Melody valve) with surgical bioprosthetic pulmonary valve replacement [30]. At 10-year follow-up, there was no difference in freedom from valve replacement between the groups.

Some data in patients with RVOT dysfunction (obstruction or regurgitation) and moderate to severe tricuspid regurgitation suggest that TPVI may cause acute reduction in the severity of tricuspid regurgitation and improvement in RV remodeling [31,32]. More data are needed on the impact of TPVI on tricuspid regurgitation.

SAPIEN TPV — A multicenter study reported outcomes for SAPIEN valve implantation in 774 patients (51 percent with a native or prior RVOT patch placement; 24 percent with a conduit; and 25 percent with a bioprosthetic valve). SAPIEN valve implantation was technically successful in 97.4 percent [33]. Valve function at discharge was excellent in most patients, but 8.5 percent had moderate or greater pulmonary regurgitation or maximum Doppler gradients >40 mmHg. During follow-up in 349 patients (median 12 months), nine patients were diagnosed with endocarditis, and 17 additional patients underwent surgical valve replacement or valve-in-valve TPVI.

For self-expanding valves — Follow-up data for self-expanding valves are limited to small short- to medium-term studies.

●Harmony TPV – A study of one-year outcomes included 87 patients who received a commercially available Harmony TPV [34]. At one year, there were no deaths and most patients had mild or less pulmonary regurgitation (98 percent of TPV22 patients, 97 percent of TPV25 patients).

●Venus-P TPV – Two multicenter follow-up studies (12 to 25 months follow-up) with a total of 93 patients showed good early results with the Venus-P TPV. Stent migration occurred in three patients, and stent frame fracture was seen in 27 percent of patients during follow-up. Most implanted valves showed excellent valve function at follow-up, and four patients had infective endocarditis [35,36].

●Pulsta TPV – A study reported successful implantation in all 25 patients receiving the Pulsta TPV for native RVOT lesions, no serious device-related adverse events, and at mean 33 months follow-up, a mean pressure gradient across the Pulsta valve of 6.5±3.0 mmHg without significant pulmonary regurgitation [37].

Initial reports of successful treatment with Harmony [34,38] and Alterra [14] are available.

ACUTE COMPLICATIONS — Acute severe complications of TPVI are uncommon. In a case series of 152 patients undergoing TPVI, six patients (4 percent) had acute complications requiring rescue surgery [39]. These included homograft rupture, dislodgement of the stented valve, occlusion of the right pulmonary artery, and compression of the left coronary artery. Data are still developing on complications for Harmony and Alterra valves.

Coronary artery compression — Compression of the left main coronary artery after device deployment occurs in approximately 0 to 1 percent of all cases of TPVI and nearly always has a fatal outcome [7,8,27,28]. Therefore, a balloon test to exclude potential coronary artery compression should always be performed prior to deployment of the valve [10]. A balloon test positive for coronary artery compression was initially identified in approximately 1 percent of cases referred to TPVI [27]. A later study found coronary compression during the balloon test in 5 percent of the patients [40]. The intervention was abandoned in these cases. (See 'Balloon test' above.)

Rarely, hematoma formation from conduit laceration may cause acute extrinsic coronary artery compression requiring emergency surgery [41].

Conduit rupture — Risk factors for conduit rupture include heavy calcification of the conduit and an RV outflow tract (RVOT) homograft valve [42].

●Contained conduit rupture may easily be treated with covered stent implantation.

●Noncontained conduit rupture is an emergency. If the rupture is not immediately controlled by covered stent implantation, emergency surgery may be necessary (less than 1 percent of patients undergoing TPVI) [7,8,27].

Ventricular arrhythmias — Cardiac telemetry is routinely performed after TPVI procedures. Limited data have raised some concern about early (within the first 24 hours) postoperative ventricular tachycardia after annular implantation of self-expanding valves [43] or the Alterra adaptative stent. It has been postulated that this risk may stem from interaction between the device and native RVOT tissue.

In addition, many patients undergoing TPVI (such as those with tetralogy of Fallot) have elevated longer term risk of ventricular arrhythmias. Since Venus-P, Harmony, and Alterra are anchored in the myocardium of the RVOT, access for electrophysiologic ablation therapy may be technically challenging or even impossible after these devices have been placed. Therefore, many centers perform electrophysiologic study prior to TPVI or surgical pulmonary valve replacement, as discussed separately. (See "Tetralogy of Fallot (TOF): Long-term complications and follow-up after repair", section on 'Patients undergoing PVR'.)

OTHER COMPLICATIONS

Stent fracture — The risk of transcatheter pulmonary valve (TPV) stent fracture varies among TPV valve types.

For Melody TPV — Stent fractures (image 4) occur in 20 to 21 percent with the Melody valve without prestenting [8,27] and in 5 to 16 percent with prestenting [7,28]. In the majority of cases, these stent fractures have no significant hemodynamic effect [44]. However, stent fractures may ultimately lead to embolization of stent parts or restenosis [27]. Risk factors for stent fracture include TPVI without prestenting (particularly with the Melody valve), severely obstructed conduits, TPVI into a native outflow tract, and presence of TPV recoil during balloon deflation [44,45].

Prestenting of the conduit with a bare metal stent before TPVI with the Melody valve is standard practice in most centers, since observational data suggest that prestenting to relieve conduit or native RV outflow tract (RVOT) obstruction may reduce the risk of stent fractures [15,45,46]. In an analysis of data from three prospective multicenter studies evaluating TPVI in patients with tetralogy of Fallot (TOF), the five-year reintervention rate among patients who had valves placed into previously stentless conduits (n = 251) was 27 percent [46]. Placement of a new prestent prior to valve implantation was associated with decreased risk of pulmonary valve prosthesis-related stent fracture (hazard ratio [HR] 0.23, 95% CI 0.10-0.55) and reintervention (HR 0.53, 95% CI 0.30-0.93).

For SAPIEN 3 TPV — Stent fractures have not been reported after TPVI with the SAPIEN 3 valve. However, in the COMPASSION study, 91 percent of the patients were prestented preceding SAPIEN implantation [12]. SAPIEN TPVI without prestenting has been performed, as illustrated by a report of 57 patients with no presenting and no stent fracture observed with median 5.7 months follow-up [16]. The rationale for this approach is that the stent of the SAPIEN valve is stronger than the stent of the Melody valve and thus may be less prone to stent fracture.

For self-expanding TPVs — Limited data are available on the risk of stent fractures for self-expanding TPVs. In a series of 89 patients receiving the Harmony TPV, all patients receiving the TPV25 were free from major stent fracture at one year. One patient with implanted with the TPV22 had a major stent fracture with valve collapse at the one-month visit, treated with surgical explantation. One-year freedom from major or minor stent fracture was 90 percent in the TPV25 group and 81 percent in the TPV22 group. Thrombus was detected in one patient with a TPV25.

Endocarditis

Risk — All patients with prosthetic valves, including those who have undergone TPVI, are considered at highest risk of adverse outcomes from infective endocarditis (IE). In addition, many patients undergoing TPVI (such as those with repaired TOF) have one or more additional risk factors for endocarditis, such as the presence of residual defects at or adjacent to the site of prosthetic material and presence of a prosthetic conduit. (See "Prevention of endocarditis: Antibiotic prophylaxis and other measures", section on 'Which patients?'.)

IE after TPVI has been observed in approximately 1 to 3 percent of patients during one- to four-year follow-up [7,8,27,28,47-51]. In a multicenter study of 2476 patients who underwent TPVR with a Melody or SAPIEN valve, IE was diagnosed in 7.3 percent of patients over a median follow-up of 2.7 years (incidence rate of 2.2 per 100 patient-years) [48]. Risk factors for IE included younger age, history of endocarditis prior to valve implantation, and higher residual gradient after TPVR. In this study, the type of valve (Melody or SAPIEN) was not an independent risk factor for IE.

The rate of IE after TPVI is similar to the generally observed rate of IE after surgical RVOT reconstruction, which ranges from approximately 0 to 5 percent of patients during one- to four-year follow-up [52-59]. However, the rate of IE after TPVI may be higher than that seen after RVOT reconstruction with a homograft. A retrospective study included 677 patients with 738 right ventricular outflow conduits implanted between 1989 and 2013 [60]. Kaplan-Meier survival free of IE for patients with surgically implanted homografts was 98.7 percent at five years and 97.3 percent at 10 years; for surgically implanted Contegra conduits, it was 87.8 percent at five years and 77.3 percent at 10 years; and for transcatheter Melody valves, it was 84.9 percent at five years. Thus, the Contegra conduit and Melody valved stents (both produced from bovine jugular vein) were associated with significantly higher incidence of IE than homografts. However, it should be noted that a Melody valve is often implanted in a stenotic homograft and some have postulated a possible role of undetected homograft IE. Thus, further studies are needed to clarify these important observations.

Among patients with surgical or transcatheter bioprosthetic pulmonary valves, the risk of endocarditis appears to be highest among those with bovine jugular vein valves. The mechanism is incompletely understood [47,60]. In a meta-analysis of 50 studies (most were observational) of >7000 patients who underwent RV-to-pulmonary artery conduit or percutaneous pulmonary valve implantation, the overall incidence of IE was 2.6 percent; the incidence was higher in bovine jugular vein valves compared with other valves (5.4 versus 1.2 percent) [61].

Diagnosis — Vegetations on a transcatheter pulmonic valve or conduit can be difficult to visualize by echocardiography [49,62]. A report suggested that intracardiac echocardiography can be helpful when transthoracic and transesophageal echocardiography fail to show vegetations [50]. Alternate imaging with cardiac CT, positron emission tomography, or CMR may also be helpful, as discussed separately. (See "Prosthetic valve endocarditis: Epidemiology, clinical manifestations, and diagnosis".)

Prevention — Preventive measures include maintenance of oral hygiene, timely treatment of infections, and antibiotic prophylaxis for bacterial endocarditis for relevant high-risk procedures (such as all oral or dental procedures that involve manipulation of gingival tissue, the periapical region of teeth, or perforation of the oral mucosa). Recommendations for antibiotic prophylaxis for endocarditis are discussed in detail separately. (See "Prevention of endocarditis: Antibiotic prophylaxis and other measures".)

REINTERVENTION — Reintervention within 5 to 10 years is required in approximately 20 to 25 percent of patients with tetralogy of Fallot undergoing TPVI [26]. Reasons for reintervention include stent fracture, device failure with recurrent stenosis, residual stenosis not relieved by the stent, or infection [23,63]. In one report, the most common reason for reintervention after TPVI was recurrent stenosis related to stent fracture [23]. As noted above, observational data suggest that prestenting may reduce the risk of stent fracture in patients receiving the Melody TPV. (See 'Melody TPV' above.)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Cardiac valve disease".)

SUMMARY AND RECOMMENDATIONS

●Use – Transcatheter pulmonary valve implantation (TPVI) is an alternative to surgical pulmonary valve replacement for selected patients with native or conduit right ventricular outflow tract (RVOT) dysfunction involving a bioprosthetic or native valve, including:

•Pulmonary regurgitation with associated criteria (table 1) and mild or no nonvalvular RVOT obstruction, as discussed separately. (See "Tetralogy of Fallot (TOF): Long-term complications and follow-up after repair", section on 'Indications'.)

•Pulmonary valve stenosis with associated criteria for intervention, which are discussed separately. (See "Tetralogy of Fallot (TOF): Long-term complications and follow-up after repair", section on 'Indications' and "Pulmonic valve stenosis in adults: Management", section on 'Indications for intervention'.)

●Exclusions – TPVI is not performed in the following settings due to technical and/or safety concerns: RVOT size that exceeds the limit of available devices, inadequate venous access from femoral and jugular sites, severe pulmonary valve stenosis that cannot be relieved by balloon dilation, risk of coronary compression by the expanded implant as determined by balloon testing, and active systemic infection.

●Devices – For patients undergoing TPVI, the choice of transcatheter pulmonary valve (TPV) is based on availability and RVOT size (with self-expanding valves used for the largest RVOTs, such as native RVOTs).

TPVs approved for use in both the United States and Europe include balloon-expandable valves (Melody and SAPIEN 3) and self-expanding valves (Harmony and SAPIEN 3 with Alterra adaptive prestent). Venus P-Valve is another self-expanding valve approved for use in Europe.

●Balloon test – Since coronary artery compression is a life-threatening complication of TPVI, a balloon test is performed in the RVOT to exclude risk of this complication (image 1). (See 'Implantation technique' above and 'Coronary artery compression' above.)

●Conduit preparation and prestenting – If there is significant conduit stenosis, conduit balloon angioplasty and stenting is performed to reduce the RVOT gradient. Prestenting of the conduit before Melody valve implantation is standard practice at most centers, since prestenting is associated with lower risk of TPV stent fractures. Prestenting is not required prior to SAPIEN valve or self-expanding valve implantation. (See 'Conduit preparation and prestenting' above.)

●Complications – The most common complication following TPVI with the Melody valve without prestenting is stent fracture, which usually is hemodynamically benign. Uncommon complications occurring in 2 percent or less of patients include clinically significant conduit rupture, coronary artery compression, ventricular arrhythmias, and endocarditis.

●Endocarditis – Patients who have undergone TPVI are considered among those at highest risk for adverse outcomes from infective endocarditis (IE). Preventive measures include antibiotic prophylaxis for invasive dental or oral procedures. (See 'Endocarditis' above and "Prevention of endocarditis: Antibiotic prophylaxis and other measures".)