Advertisement

Team Approach to Decision-Making in Pulmonary Valve Replacement

      Unlabelled image
      Algorithm for the evaluation of a patient with RVOT (right ventricular outflow tract) dysfunction.
      Central Message
      Percutaneous PVR is may be preferred for isolated pulmonary valve disease with appropriate anatomy and surgery is preferred when concomitant abnormalities present. The technique of surgical PVR should facilitate future transcatheter valve-in-valve treatment options.

      Introduction

      Isolated pulmonary valve replacement (PVR) or right ventricle-pulmonary artery (RV-PA) conduit replacement is the most common late intervention following repair of congenital heart disease (CHD), usually tetralogy of Fallot or other conotruncal anomalies. Recent advances in transcatheter technology to manage the right ventricular outflow tract (RVOT) demand a collaborative approach between surgeons and cardiologists as part of a multidisciplinary team with systematic review of each patient and shared decision-making to determine the optimal treatment plan. While the indications for PVR, imaging and timing related to PVR, and outcome data of both surgical (sPVR) and transcatheter PVR (tPVR) are detailed elsewhere in this issue, this article focuses on the considerations, collaborative teamwork and decision-making between cardiology and surgery as to the optimal approach for a given patient. In general, rigid prescriptive algorithms are avoided and patient care is individualized. The purpose of this review is to outline factors to be considered and how those influence shared decision making.

      The importance of a team

      Few areas of cardiac surgery rely as heavily on colleagues in multiple specialties as does the diagnosis and treatment of congenital heart disease. The first important consideration in building a successful team-based program is having engaged surgeons and cardiologists and other specialists with expertise in heart failure, imaging (echocardiography, cross-sectional imaging, and advanced imaging modalities when appropriate), diagnostic and therapeutic cardiac catheterization, and electrophysiology. In addition, pre- and postprocedure (interventional or surgical) management of the patient should also include risk assessment for sudden cardiac death (SCD), cardiac rhythm surveillance and medical management. Many congenital heart defects result in the need for repeated interventions or surgical procedures over a lifetime. Since there is evidence that late survival is decreased when a patient reaches the fifth sternotomy,
      • Holst K.A.
      • Dearani J.A.
      • Burkhart H.M.
      • et al.
      Risk factors and early outcomes of multiple reoperations in adults with congenital heart disease.
      the ideal surgical strategy should be to coordinate a programmatic approach that includes a combination of periodic percutaneous interventions when feasible so that a patient never reaches the sternotomy operation in their lifetime. Ideally, these discussions occur during a regularly scheduled, multidisciplinary conference so all parties involved can review the work-up, imaging, and voice their analysis and opinion to the other members of the multidisciplinary team and come to an agreement as to the best plan moving forward (Fig. 1).
      Figure 1
      Figure 1Algorithm for the evaluation of a patient with RVOT (right ventricular outflow tract) dysfunction. Color version of figure is available online.

      PVR - Relevant Considerations for the Interventionalist and Surgeon

      There are important considerations that play into the decision to recommend either a percutaneous procedure or surgery…some anatomic or technical, and some clinical and physiologic (Fig. 2). Anatomic and technical factors to be considered are outlined below with some caveats on how these can sway the decision-making toward surgery or a transcatheter approach. In general, when multiple abnormalities exist, such as example, pulmonary regurgitation, tricuspid regurgitation (TR), residual ventricular septal defect (VSD), we recommend surgical intervention with the surgical plan to address all areas of pathology, or at least the most hemodynamically significant ones. In cases of a single pathologic lesion, such as obstruction and/or regurgitation of the RVOT that can be completely addressed with a transcatheter approach, an interventional approach might be favored when anatomy is appropriate. Exceptions are also made for patients who are considered poor surgical candidates. However, each patient is considered individually in the context of a multidisciplinary evaluation as outlined above and shared decision-making between the specialties and the patient/family is applied.
      • 1)
        Current RVOT anatomy: Patients may have native pulmonary valve dysfunction, prosthetic valve dysfunction (stented, stentless or homograft), or RV-PA conduit dysfunction. If a bioprosthetic valve is in place, assessment is performed to determine if the diameter is adequate for percutaneous valve-in-valve (V-in-V) therapy with an adequately sized percutaneous valve. It may be feasible to perform a frame fracture of the prosthesis to improve the inner diameter, but some prosthetic valves and valved conduits, such as Hancock, cannot be fractured. Other conduits may be severely calcified or of a material that does not allow expansion, potentially limiting the final caliber of the percutaneously implanted valve. In addition, individuals with RV-PA conduits frequently have multilevel obstruction and a tPVR might be inadequate to address all levels of obstruction. There are situations when a transcatheter approach in younger children is sufficient to relieve a gradient, but a surgical approach allowing a substantially larger bioprosthetic valve may be preferred to facilitate future transcatheter interventions. For patients with structural abnormalities of the RVOT and main pulmonary artery (PA), for example, multilevel obstruction, pseudoaneurysm, etc., we prefer surgery that would provide a more definitive long-term solution while minimizing large amounts of indwelling hardware.
      • 2)
        Position of conduit and relevant adjacent structures: The conduit may be orthotopic (such as after a Ross procedure) or nonorthotopic; that is, extracardiac (such as in pulmonary atresia-VSD, TGA-VSD, truncus arteriosus, etc.). The position of the conduit may put adjacent structures at risk, most commonly the left main coronary artery posteriorly, especially after Ross procedure, or the left anterior descending artery (LAD) lateral to an extracardiac conduit or to a RVOT patch. Consideration should also be made of anomalous coronaries, such as LAD arising from the right coronary artery. The sternum or chest wall may also be a limiting factor in being able to provide a patent RVOT, but if a surgical approach is performed, the underside of the sternum in the area where the prosthesis needs to sit can be shaved or resected allowing adequate room for the prosthesis. Preoperative CT angiography is essential for complete assessment of coronary artery anatomy for both the surgeon and the interventionalist.
      • 3)
        Pulmonary artery abnormalities: Many patients have pulmonary artery pathology that can range from areas of hypoplasia, stenosis, dilation, or aneurysm and may have had multiple various interventions including patches and stents. Frequently the RPA can be hypoplastic or compressed as it courses posterior to the aorta, especially in cases of aortic dilation. Ostial or proximal branch PA stenosis is easily addressed at the time of sPVR with patch angioplasty. Stenosis of the right PA may be treated with stenting when aortic diameter is normal or only mildly dilated, whereas significant aortic dilation with right PA stenosis is better treated with ascending aortic replacement and patch angioplasty of the right PA. Left PA stenosis is best treated percutaneously since surgical access can be difficult in a reoperation and the phrenic nerve may be at risk with mediastinal dissection; we have also used a hybrid approach with intraoperative stent placement under direct visualization at the time of sPVR. There may be cases in which a “kinking” of the LPA has occurred with RV enlargement that may improve with RV remodeling after PVR. Aggressive percutaneous stenting of mild proximal branch PA should be avoided as indwelling stents almost always makes surgery in stented areas difficult when reoperation is needed. PA aneurysms are also managed surgically with resection or plication maneuvers. Pathology beyond the orifice of the left pulmonary artery or distal right PA (ie, beyond pericardial reflection) is difficult to access surgically in a redo mediastinum; percutaneous strategies are generally preferred in those settings.
      • 4)
        Concomitant hemodynamic lesions/pathology: Patients may have residual or recurrent atrial or ventricular level shunts or valve lesions, most commonly regurgitation. The degree of valve pathology, regurgitation or stenosis, and the decision to intervene on the valve (usually valvuloplasty and/or annuloplasty) should be determined preoperatively since the conditions of anesthesia downgrade and underestimate the hemodynamic significance. Presence of residual intracardiac shunts at any level (atrial or ventricular) should be determined preoperatively, particularly in the setting of a hazardous re-sternotomy that may require the institution of bypass to facilitate the re-entry and avoid air embolism. In general, we aim to close all residual shunts at the reoperation, in part for safety on the next potentially hazardous re-sternotomy. Other pathology to look for include constrictive pericarditis or coronary artery disease. Although very uncommon, the clinical findings of constriction are similar to right-sided heart failure and so should always be in the differential diagnosis. If constriction is suspected clinically, identified by echocardiography, and confirmed by cross-sectional imaging and cardiac catheterization, then pericardiectomy should be performed as a concomitant procedure.
      • 5)
        Associated atrial and ventricular arrhythmias: Many of these patients, particularly those with tetralogy of Fallot, have concomitant arrhythmias. Atrial tachyarrhythmias (flutter and fibrillation) are most common and can be treated either percutaneously or surgically. Ventricular arrhythmias may also occur, particularly in patients who have had a right ventriculotomy. The need for preoperative electrophysiology study for diagnosis and/or treatment is essential. Depending on the findings of the preoperative electrophysiology study, a percutaneous ablation may be done during that study, and may also require additional ablation lesions to be performed intraoperatively. At other times, it may be preferred to do a surgical ablation. The need for permanent pacing or implantable cardioverter/defibrillator (ICD) placement is also not uncommon. The importance of a close collaboration with the electrophysiology team to determine the best approach in these patients cannot be overemphasized. The treatment strategy may include percutaneous intervention (ie, ablation), surgical ablation, or both.
      • 6)
        Tricuspid valve and indwelling pacemaker/ICD leads: In general, we are proactive regarding intervention on the tricuspid valve (TV) and prefer a surgical approach when TR is present, particularly when there is important structural TR (as opposed to functional TR) or in the setting of indwelling pacemaker/ICD leads. Although the degree of TR generally improves following both tPVR and sPVR, as the pulmonary valve eventually becomes dysfunctional, the degree of TR inevitably worsens, and empiric tricuspid repair may delay the timing of the next intervention. For this reason, when there is ≥moderate TR or the annulus is ≥4 cm noted preoperatively, we favor a surgical approach with the intent to repair the TV at the time of sPVR and make a concerted effort to preserve the transvenous pacing leads if feasible. Transvenous leads can be repositioned at a commissure to reduce lead-induced TR or can be removed and an epicardial pacing system placed at the time of sPVR.
        • Saran N.
        • Dearani J.A.
        Tricuspid valve repair: How I teach it.
        For those with a new pacing requirement undergoing sternotomy, we prefer placement of an epicardial system, particularly in younger patients to avoid potential injury to the TV in the long-term.
      • 7)
        Aortic and aortic valve pathology: It is common for patients with conotruncal anomalies to have aortic dilation but without the same risk of adverse events as documented in acquired diseases of the ascending aorta and aortic valve.
        • Stulak J.M.
        • Dearani J.A.
        • Burkhart H.M.
        • et al.
        Does the dilated ascending aorta in an adult with congenital heart disease require intervention?.
        ,
        • Dearani J.A.
        • Burkhart H.M.
        • Stulak J.M.
        • et al.
        Management of the aortic root in adult patients with conotruncal anomalies.
        In our practice we plan ascending aortic replacement at 55 mm at the time of sPVR. When the aortic dilation is the main driver for intervention, we wait until the aortic diameter exceeds 6 cm but intervene earlier if concomitant significant aortic valve disease (stenosis or regurgitation) is present. In the setting of aortic dilation and aortic regurgitation in a patient who prefers a valve-sparing aortic root replacement, we intervene before the regurgitation becomes severe to optimize the late durability of a valve sparing aortic root replacement.
      • 8)
        Coronary artery disease: Adults with congenital heart disease can present with concomitant atherosclerotic coronary artery disease and require equivalent preoperative evaluation to adults with acquired cardiac disease undergoing cardiac surgery. Since most patients will undergo a pre-procedural CT angiography for mediastinal anatomic delineation, we use that CT to also evaluate for the presence of coronary atherosclerosis. If CT angiography is abnormal, then conventional coronary angiography is required to further assess and plan any coronary intervention – either PCI or CABG. The management of coronary artery disease is individualized. In general, we use the left internal mammary artery to graft LAD disease given the survival benefit it confers,
        • Dearani J.A.
        • Burkhart H.M.
        • Stulak J.M.
        • et al.
        Management of the aortic root in adult patients with conotruncal anomalies.
        but often prefer a hybrid approach and percutaneous stent placement for circumflex and right coronary artery lesions depending on the complexity of the reoperation. This is due in part to the presence of significant cardiomegaly that is often present, mediastinal adhesions that may be hostile, simplifying future reoperation with fewer grafts in the area of the RVOT, and minimizing left phrenic nerve injury which may also be at risk with right-sided pulmonary valve/conduit surgery with left posterolateral wall dissection. It should be highlighted that data on prognostic implications and benefits of coronary revascularization (surgical or percutaneous) in adults with congenital heart disease are very limited.
      • 9)
        Redo-sternotomy: As referenced above, the number of previous sternotomies has been shown to impact late survival.
        • Holst K.A.
        • Dearani J.A.
        • Burkhart H.M.
        • et al.
        Risk factors and early outcomes of multiple reoperations in adults with congenital heart disease.
        This should be taken into account along with how hazardous the re-entry appears and vascular access for potential peripheral cannulation during or before re-entry. Unfortunately, some patients who have had many prior interventions, particularly when they were young, may have limited peripheral vascular access, which in some instances can further complicate safe strategies for re-entry and increase the risk of reoperation. The use of protective membranes placed during previous operations has been applied in many programs and can protect vital structures during the sternal re-entry, but those structures can remain at risk during the mediastinal dissection after the sternum is divided. In general, we use Gore-Tex pericardial membrane to cover either an extracardiac conduit or aorta if proximity or adherence to the sternum is anticipated.
      • 10)
        Age and size of pediatric patients: For some pediatric patients, particularly those who have undergone RV-PA conduit placement starting in the neonatal period, their clinical course requires relatively frequent interventions. For these younger patients, consideration should be made regarding optimizing a combined surgical and percutaneous strategy. At various time points in their care, a temporizing palliative strategy with RVOT +/- PA stenting and sequential balloon angioplasty may allow sufficient somatic growth prior to the next surgical intervention, allowing a larger pulmonary bioprosthesis (or RV-PA conduit) to be placed that can then facilitate future tPVR(s). Otherwise, the limitations of tPVR in young children are primarily related to the caliber of the femoral or jugular veins (to accommodate valve delivery). In general, children >20 kg are likely to have adequate vascular access for tPVR, and tPVR has been reported in even smaller patients.
        • Martin M.H.
        • Shahanavaz S.
        • Peng L.F.
        • et al.
        Percutaneous transcatheter pulmonary valve replacement in children weighing less than 20 kg.
        Such scenarios highlight the need for close collaboration and planning between surgical and interventional teams.
      • 11)
        Risk of infective endocarditis: Infective endocarditis is a rare but serious complication following sPVR or tPVR. Registry data have recently demonstrated an annualized incidence of 2.2 per 100 patient years after tPVR with risk factors for infective endocarditis including younger age, history of prior infective endocarditis, and residual gradient across the implanted valve.
        • McElhinney D.B.
        • Zhang Y.
        • Aboulhosn J.A.
        • et al.
        Multicenter study of endocarditis after transcatheter pulmonary valve replacement.
        In our practice, patients with an episode of treated prosthetic valve endocarditis traditionally undergo sPVR. Similarly, in a patient with a small prosthesis or conduit, or those with multilevel obstruction, surgery may be the preferred approach to minimize residual gradient across the implanted valve. Meticulous sterile technique in both the catheterization lab and operating room cannot be overemphasized to minimize the potential risk of endocarditis during PVR. Pre-procedural antibiotic prophylaxis is generally recommended in the setting of a prosthetic heart valve.
      • 12)
        Clinical status: the clinical condition of the patient and other physiologic and medical factors that also affect the treatment pathway include:
        • 1)
          comorbidities – that is, pulmonary, renal, and hepatic disease most commonly; for those with substantial chronic kidney disease or intrinsic pulmonary disease requiring home oxygen, a percutaneous strategy is commonly favored. Obesity is a relative contraindication for resternotomy.
        • 2)
          cardiac risk factors – that is, right, or left ventricular systolic dysfunction; pulmonary hypertension; atherosclerotic coronary artery disease
        • 3)
          functional status, particularly those in class IV heart failure. Since isolated pulmonary pathology should not result in overt right heart failure, it is critical to look for concomitant hemodynamic and structural lesions (severe TR, constriction, severe right ventricular systemic dysfunction, restrictive cardiomyopathy, etc.) in these patients.
      Figure 2
      Figure 2Summary of key factors to be considered and how those influence the decision toward or away from surgery. TR, tricuspid regurgitation; RVOT, right ventricular outflow tract; TPVR, transcatheter pulmonary valve replacement. Color version of figure is available online.

      Surgical PVR Pathway

      In general, our approach to sPVR, either in the native pulmonary artery or in the setting of a conduit, is to place a stented bioprosthesis in the native PA or conduit pathway with an intact back wall of the conduit bed and roof over the anterior conduit defect with bovine pericardium. We aim for a 27 or 29 mm porcine or pericardial bioprosthesis in an adult. When a previous transannular patch has been used, we resect much of the redundant patch over the RVOT. Surgical strategies to help facilitate and sustain RV “remodeling” include (1) using a large prosthesis to reduce the RV pressure to (near) normal, (2) resection of a redundant RVOT patch to minimize RV dyskinesis, and (3) proactive tricuspid repair to minimize residual/recurrent TR in the long term.
      In our experience, there has been no demonstrable difference in durability between porcine or pericardial valves;
      • Egbe A.C.
      • Connolly H.M.
      • Miranda W.R.
      • et al.
      Outcomes of bioprosthetic valves in the pulmonary position in adults with congenital heart disease.
      this has also been noted by others. Our experience with pulmonary or aortic homografts in the pulmonary position in adolescents and adults has demonstrated poor durability, therefore homografts are generally avoided in this older age group. In fact, we begin to use stented bioprosthetic valves for sPVR in children beginning at approximately 5 years of age since roof augmentation of the RV-PA pathway allows a large prosthesis to be placed. There are no absolute guidelines with regarding sPVR or tPVR intervention relative to age or size. In general, when a 23 mm bioprosthetic valve can be safely implanted with or without transannular patch augmentation, we lean towards surgery. Besides the relatively good durability of bioprosthetic pulmonary valves in children and adults, the interventional cardiologist prefers a stented bioprosthesis to facilitate future “valve-in-valve” tPVR, since the landing zone is secure and reduces the possibility of adjacent coronary artery compression. Proximal branch PA stenosis is treated by patch angioplasty at the time of sPVR. Distal PA stenosis is treated by hybrid stent placement or intraoperative dilation of a pre-existing stent.
      Calcification can be particularly challenging, but with endarterectomy of calcified areas in locations of suture placement, we have been able to achieve proper suturing/seating of the valve. The ideal prosthesis position is closer to the pulmonary confluence with the struts pointed posteriorly toward the confluence. This more distal positioning also keeps the prosthesis away from the left main coronary artery posteriorly and the LAD laterally. If the decision is to replace the conduit entirely, we build a “bio-Bentall conduit” with a bioprosthetic valve sutured in between 2 segments of Dacron tube graft, and position the prosthesis closer to the confluence as described above. A mechanical prosthesis is rarely used because of the relatively good durability of bioprosthetic valves in the pulmonary position and feasibility of future tPVR. However, if a mechanical valve is used because of the need for warfarin anticoagulation for other reasons, for example, left-sided mechanical valve, or accelerated bioprosthetic valve deterioration, then the target INR should be 2.5–3.5.
      When other abnormalities are present beyond pulmonary valve dysfunction, for example, TR and residual VSD, we prefer a surgical approach with the intent that all abnormalities will be addressed (Fig. 3). Our approach to TR is aggressive and proactive – tricuspid repair for moderate regurgitation noted on preoperative transthoracic echocardiogram, an annulus of ≥40 mm, or elevated pulmonary artery pressures. While we recognize that correction of pulmonary valve pathology will result in RV remodeling and can improve the TR, we believe that maintenance of tricuspid competency is optimized in the long term when recurrent pulmonary valve dysfunction occurs with subsequent RV dilation or increased RV pressure that may help delay the next intervention triggered by clinical right-sided heart failure. Management of patients with lead-induced TR is individualized and can include tricuspid repair or replacement or removal of the transvenous system and placement of an epicardial system.
      • Saran N.
      • Dearani J.A.
      Tricuspid valve repair: How I teach it.
      The presence of pseudoaneurysm along the RV-PA pathway requires surgical repair; rarely would a percutaneous approach be utilized for this abnormality.
      Figure 3
      Figure 3Summary of surgical considerations in addition to PVR (pulmonary valve replacement). PPM/AICD, permanent pacemaker/automatic implantable cardioverter/defibrillator; intraop, intraoperative; AR, aortic regurgitation; VSRR, valve-sparing aortic root replacement; mod-severe, moderate-severe; RVOT, right ventricular outflow tract; LIMA, left internal mammary artery; LAD, left anterior descending; PCI, percutaneous cardiac intervention. Color version of figure is available online.
      Atrial or ventricular arrhythmias are frequently present in this group of patients and often require an electrophysiologic evaluation prior to any percutaneous or surgical intervention. Results of a preoperative electrophysiology study will direct percutaneous therapy prior to a percutaneous intervention or intraoperative surgical treatment – right sided vs biatrial maze, cavotricuspid isthmus ablation, or ablation of the RVOT for ventricular arrhythmias (often at the superior aspect of the VSD patch at the infundibulum). The need for permanent pacing or need for ICD placement is also determined by a thorough electrophysiology evaluation.

      Transcatheter Treatment Pathway

      Current RVOT anatomy

      Circumferential Conduit

      Considerations when evaluating for a potential tPVR are shown in Figure 4. The 1st FDA approved transcatheter pulmonary valve, the Melody valve by Medtronic, was designed specifically for use in a RV-PA conduit. Clinical experience has shown that conduits in general, and homograft conduits in particular, are prone to significant calcification and stenosis. This progressive conduit deterioration does not preclude transcatheter pulmonary valve implantation, but it does necessitate coronary compression testing and pre-tPVR conduit rehabilitation; that is, stent dilation to create a valve “landing zone” of adequate size without compromise of the left coronary artery or anomalous LAD and potential need for covered stents if significant dilation is performed. Whereas the initial Melody valve experience involved primary implantation of the Melody valve without “pre-stenting,” a significant rate of Melody valve stent fracture and subsequent valve dysfunction was noted in such situations. Thus, current practice involves bare metal and/or covered stent implantation prior to valve deployment to improve the valve stent frame. The Edwards Sapien valve does not carry the same risk of stent fracture as the Melody valve, but pre-stenting may still be needed to treat conduit stenosis.
      Figure 4
      Figure 4Considerations for transcatheter pulmonary valve replacement (TPVR). PS/PR, pulmonary stenosis/pulmonary regurgitation; CTA, computed tomography angiogram; PVR, pulmonary valve replacement. Color version of figure is available online.

      Bioprosthetic Valve (Including Ringed-valve Conduit, for example, Hancock)

      “Valve-in-valve” tPVR is a generally straightforward, low risk procedure that has been shown to effectively treat bioprosthetic pulmonary valve regurgitation and/or stenosis. The bioprosthetic valve structure provides an ideal landing zone for the transcatheter valve and protects surrounding structures from impingement by the balloon-expandable valve. Radiopaque features of the surgical valve housing facilitate valve positioning using fluoroscopy.
      A small surgical bioprosthesis may present a challenge for tPVR. A transcatheter valve implanted into a 21 or 23 mm bioprosthetic valve will likely not provide adequate effective orifice area (EOA) for an average sized adult. The surgical ring of some bioprosthetic valves can be fractured with high-pressure dilation balloons prior to tPVR to improve the final EOA after valve implantation. However, some bioprosthetic valves, particularly the Hancock, are unable to be fractured with currently available high-pressure balloon technology and may limit an appropriate-sized tPVR.

      Native RVOT (Including Operated—for example, Transannular Patch, Valvotomy/Valvectomy/Valvuloplasty)

      The absence of circumferential prosthetic material in the RVOT does not preclude transcatheter valve-in-valve. In a native RVOT of smaller diameter, Melody valve implantation (with supportive pre-stenting) may be feasible. A larger RVOT may be amenable to Sapien valve implantation in an off-label indication, typically without pre-stenting. To determine eligibility, a large sizing balloon, often 35 mm in diameter, is inflated across the RVOT to determine the minimum diameter of the outflow tract with simultaneous ascending aorta angiogram to assess for coronary artery compression or aortic valve deformation and new regurgitation. If the balloon waist is <31 mm with no impingement on surrounding structures, a Sapien valve may be safely implanted.
      More recently, the Harmony valve (Medtronic, Minneapolis, MN) has been approved for implantation into the native RVOT. Unlike the balloon-expandable Melody and Sapien valves, the Harmony valve is a self-expanding porcine pericardial valve with an hourglass shaped frame specifically designed for larger, native RVOT use, such as in the setting of a transannular patch after repair of tetralogy of Fallot. Patient screening for Harmony valve involves a gated CT angiogram with measurement of systolic and diastolic dimensions of the dynamic RVOT. This analysis ensures adequate space and length are present with frame apposition to the RVOT and main PA throughout the cardiac cycle. Valve implantation is performed in a pre-determined desired position utilizing optimal fluoroscopic angles derived from the CT angiogram.

      Coronary artery compression

      An infrequent but potentially catastrophic risk of balloon-expandable valves is compression of adjacent coronary arteries; specifically, the left main or anomalous LAD posteriorly (in the setting of a conduit), the LAD laterally, or the RCA medially. If tPVR is likely to involve significant enlargement of the RVOT with pre-stenting or valve implantation, pre-procedural CT angiography is essential to identify coronary artery anatomy and their relationship to adjacent structures. In addition, coronary compression testing is performed prior to valve implantation to ensure that the RVOT can be safely dilated without coronary compromise.

      Non-cardiac considerations

      Other factors considered when deciding whether to proceed with tPVR and what type of transcatheter valve to implant, include patient size and vascular access options for valve delivery. In general, our practice is to reserve Melody valves for smaller patients, as the maximum external diameter of a Melody valve is 24–26 mm. Being bovine jugular vein, the EOA is larger than manufactured valves of the same implant diameter. The Sapien valve, by comparison, can be implanted up to 30–31 mm external diameter, making it suitable for larger patients who benefit from a larger EOA. The valve is comprised of bovine pericardium. The Harmony valve houses either a 22 mm (TPV 22) or a 25 mm (TPV 25) porcine pericardial valve within a self-expanding nitinol frame. The large, self-expanding frame is optimal for a large, aneurysmal RVOT. The flexible frame can then allow for over-expansion and subsequent valve-in-valve therapy when needed.
      Transcatheter valves are most commonly implanted from a femoral venous approach. The Melody valve delivery system (Ensemble II) hosts a 22 French outer diameter and can be advanced into position primarily. The DrySeal sheath by W.L. Gore is now utilized ubiquitously for Sapien valve implantation as it protects the TV during valve implantation. A 23, 26, or 29 mm Sapien S3 valve is mounted on the standard delivery system (Commander) and advanced through a 22, 24, or 26 French DrySeal sheath, respectively. In young patients, ultrasound-guided access allows for vessel diameter assessment prior to sheath selection with optimal sheath sizing being at or below the measured vein diameter. In the setting of a challenging catheter course or a small patient, delivery via an internal jugular vein approach may be preferred. In the absence of femoral venous or jugular venous access, our approach has been to favor sPVR, although one may consider alternative vascular approaches, including off-pump, hybrid per-ventricular tPVR.
      • Sosnowski C.
      • Matella T.
      • Fogg L.
      • et al.
      Hybrid pulmonary artery plication followed by transcatheter pulmonary valve replacement: Comparison with surgical PVR.

      Outcomes for tPVR

      One particular limitation of tPVR is the lack of outcome data, especially in the younger pediatric population. In vitro, laboratory studies demonstrate that the Sapien valve is durable out to an equivalent of 5 years of implantation.
      • Sathananthan J.
      • Hensey M.
      • Landes U.
      • et al.
      Long-term durability of transcatheter heart valves: Insights from bench testing to 25 years.
      A tPVR study from Germany with a median follow-up of 3.9 years included a combination of Melody (n = 220) and Sapien valves (n = 16) in a cohort with a median age of 18 years. In their analysis, 7% had valve failure and a postvalve implantation RVOT gradient >15 mm Hg was a risk factor for both death and valve failure.
      • Georgiev S.
      • Ewert P.
      • Tanase D.
      • et al.
      A low residual pressure gradient yields excellent long-term outcome after percutaneous pulmonary valve implantation.
      Another study from that group comparing Melody to sPVR with a median follow-up of 5.4 years showed a trend of higher rates of endocarditis in the Melody group (1.6%/year vs 0.5%/year, P = 0.08), without differences in survival or freedom from reintervention in the limited follow-up period.
      • Georgiev S.
      • Ewert P.
      • Eicken A.
      • et al.
      Munich comparative study: prospective long-term outcome of the transcatheter melody valve versus surgical pulmonary bioprosthesis with up to 12 years of follow-up.
      The 3-year results of the COMPASSION multicenter trial for Sapien implantation in the RVOT consisting of 57 patients with a median age of 27 years reported a freedom from reintervention of 94% and freedom from endocarditis of 97% at 3 years.
      • Kenny D.
      • Rhodes J.F.
      • Fleming G.A.
      • et al.
      3-Year outcomes of the Edwards SAPIEN transcatheter heart valve for conduit failure in the pulmonary position from the COMPASSION multicenter clinical trial.
      The experience of Sapien transcatheter valve replacement for adult acquired aortic valve disease can be considered regarding potential durability, although it is unclear how this can be extrapolated to the pulmonary position with different anatomy and hemodynamics, and much younger patients. While there is limited data regarding Sapien valve placement, there is much less data on Harmony regarding durability. The early feasibility results which included 17 patients that had adequate follow-up, found the mean gradient to be 16 mm Hg and that 2 patients had increased gradients, one with moderate-severe regurgitation.
      • Benson L.N.
      • Gillespie M.J.
      • Bergersen L.
      • et al.
      Three-year outcomes from the harmony native outflow tract early feasibility study.

      Follow-up Post PVR

      In our practice, anticoagulation with a vitamin K antagonist is typically recommended for 3 months following sPVR or tPVR in adolescents and adults, as postoperative anticoagulation been associated with lower rates of pulmonary prosthesis dysfunction.
      • Egbe A.C.
      • Connolly H.M.
      • Miranda W.R.
      • et al.
      Outcomes of bioprosthetic valves in the pulmonary position in adults with congenital heart disease.
      We aim for an INR of 2–2.5. One month following the discontinuation or warfarin, a follow-up echocardiogram is performed to reassess prosthetic valve function and leaflet thickness/mobility. As newer investigational transcatheter technology is introduced, postprocedure imaging is performed as per the specific investigational protocol. Lifelong low dose aspirin (81 mg) is recommended in the absence of any contraindications. In the presence of clinical and echocardiographic stability, most patients post-PVR are seen on an annual basis with echo-Doppler reevaluation. In our practice, we do not perform routine cross-sectional imaging post-PVR. We do recommend endocarditis prophylaxis and regular dental care.

      Challenges and Future Directions

      Given the reduced morbidity associated with percutaneous interventions, it is natural for patients and providers to favor non-surgical approaches. However, it is essential to gather data regarding the late durability of percutaneous prostheses, as well as the impact of patient-prosthesis mismatch and residual RVOT obstruction with regards ventricular remodeling and clinical outcomes, particularly as we consider multiple tPVR procedures to avoid or delay repeat surgery. Lastly, the risk and challenges of surgical interventions in a patient with prior tPVR need to be investigated. The latter is particularly pertinent given the growing body of literature describing the increased morbidity and mortality in patients undergoing surgical explantation of previously percutaneous aortic valve prostheses.

      REFERENCES

        • Holst K.A.
        • Dearani J.A.
        • Burkhart H.M.
        • et al.
        Risk factors and early outcomes of multiple reoperations in adults with congenital heart disease.
        Ann Thorac Surg. 2011; 92 (discussion 129-130): 122-128
        • Saran N.
        • Dearani J.A.
        Tricuspid valve repair: How I teach it.
        Ann Thorac Surg. 2018; 105: 675-679
        • Stulak J.M.
        • Dearani J.A.
        • Burkhart H.M.
        • et al.
        Does the dilated ascending aorta in an adult with congenital heart disease require intervention?.
        J Thorac Cardiovasc Surg. 2010; 140 (discussion S86-91): S52-S57
        • Dearani J.A.
        • Burkhart H.M.
        • Stulak J.M.
        • et al.
        Management of the aortic root in adult patients with conotruncal anomalies.
        Seminars in thoracic and cardiovasc surg. Pediatric cardiac surg annual. 2009; : 122-129
        • Martin M.H.
        • Shahanavaz S.
        • Peng L.F.
        • et al.
        Percutaneous transcatheter pulmonary valve replacement in children weighing less than 20 kg.
        Circ Cardiovasc Interv. 2018; 91: 485-494
        • McElhinney D.B.
        • Zhang Y.
        • Aboulhosn J.A.
        • et al.
        Multicenter study of endocarditis after transcatheter pulmonary valve replacement.
        J Am Coll Cardiol. 2021; 78: 575-589
        • Egbe A.C.
        • Connolly H.M.
        • Miranda W.R.
        • et al.
        Outcomes of bioprosthetic valves in the pulmonary position in adults with congenital heart disease.
        Ann Thorac Surg. 2019; 108: 1410-1415
        • Sosnowski C.
        • Matella T.
        • Fogg L.
        • et al.
        Hybrid pulmonary artery plication followed by transcatheter pulmonary valve replacement: Comparison with surgical PVR.
        Circ Cardiovasc Interv. 2016; 88: 804-810
        • Sathananthan J.
        • Hensey M.
        • Landes U.
        • et al.
        Long-term durability of transcatheter heart valves: Insights from bench testing to 25 years.
        JACC Cardiovasc Interv. 2020; 13: 235-249
        • Georgiev S.
        • Ewert P.
        • Tanase D.
        • et al.
        A low residual pressure gradient yields excellent long-term outcome after percutaneous pulmonary valve implantation.
        JACC Cardiovasc Interv. 2019; 12: 1594-1603
        • Georgiev S.
        • Ewert P.
        • Eicken A.
        • et al.
        Munich comparative study: prospective long-term outcome of the transcatheter melody valve versus surgical pulmonary bioprosthesis with up to 12 years of follow-up.
        Circ Cardiovasc Interv. 2020; 13e008963
        • Kenny D.
        • Rhodes J.F.
        • Fleming G.A.
        • et al.
        3-Year outcomes of the Edwards SAPIEN transcatheter heart valve for conduit failure in the pulmonary position from the COMPASSION multicenter clinical trial.
        JACC Cardiovasc Interv. 2018; 11: 1920-1929
        • Benson L.N.
        • Gillespie M.J.
        • Bergersen L.
        • et al.
        Three-year outcomes from the harmony native outflow tract early feasibility study.
        Circ Cardiovasc Interv. 2020; 13e008320