Example Of Dissertation On Follow Up And Judgment Of Bovine Pericardial Heart Valve As A Pulmonary Valve Replacement

Selection from an 8 year-period of pulmonary valve replacement and cross sectional follow up of a commercially available biological valve, assessed with established and a new method

What This Thesis Is About

Pulmonary valve replacement is a procedure necessary in many congenital heart defects and in-valve degeneration cases. Surgical results are excellent, but the longevity of artificial valves is very limited. Our goal was to give a retrospective overview of methods and results from the year 2000 until 2008 at Rikshospital, as well as the prospective five year follow-up of a biological valve made from bovine pericardium. Follow up included clinical examination with ECG, echocardiography, spirometry, treadmill testing, echocardiography under workload and MRI in some patients. We also tested a new echocardiographic based method that could provide better patient follow-up. The first article included in this thesis gives a summary of the main replacement valve types used and discusses indications as well as strength and weaknesses of the different approaches. The second article is a follow-up from a selection of patients with the Perimount valve, five years after the last operation. In the third article we compare VentriPoint examinations with the reference method MRI.

Acknowledgements

Abbreviations
ABO – blood group system
ACC – American College of Cardiology
ACSM – American College of Sports Medicine
AHA – American Heart Association
AV – Aortic valve
CHD – congenital heart defect
DPH – decellularized pulmonary homograft
ECG – electrocardiography
EDV – end diastolic volume
EF – ejection fraction
ESC – European Society of Cardiology
ESV – end systolic volume
HLA-DR – human leukocyte antigen-DR
KBR – knowledge-based reconstruction
LA – left atrium
LV – left ventricle

NYHA – New York Heart Association

PR – pulmonary regurgitation
PV – pulmonary valve
PVR – pulmonary valve replacement
RA – right atrium
RV – right ventricle
RVOT – right ventricular outflow tract
RVOTO – right ventricular outflow tract obstruction
SIS-ECM – small intestinal submucosal extracellular matrix
QRS – ECG waves Q, R, and S
TEPV – tissue-engineered pulmonary valves
TOF – Tetralogy of Fallot
TGA – transposition of the great arteries
TR – tricuspid regurgitation
VMS – VentriPoint Medical System

1. Introduction

The normal human heart is comprised of four chambers, two atriums and two ventricles, which are separated by the atrial septum and the interventricular septum (1). The heart has a total of four valves that have a critical role in assuring forward flow and preventing backward flow. Two valves are located between the atria and the ventricles. Two more valves are located between the ventricles and the great arteries. The deoxygenated blood is returned to the right atrium (RA), and the tricuspid valve passes that blood into the right ventricle (RV). When atrial pressure increases over the pressure in the ventricles, the valves between the atria and the ventricles open to pass the blood into the ventricles. The pulmonary valve is located between the RV and the pulmonary artery and is passed by the deoxygenated blood on its way to the lungs where it is oxygenated. The mitral valve is located between the left atrium (LA) and the left ventricle (LV) and allows the passage of oxygenated blood from the lungs. The aortic valve is located between the LV and the aorta and is passed by the oxygenated blood on its way to the body.
Congenital heart defects (CHD) and other causes can lead to dysfunctional processes in which surgical procedures are warranted. Pulmonary valve replacement (PVR) is a surgical procedure that addresses the dysfunctional RV to pulmonary connection, which can occur because of a right ventricular outflow tract obstruction (RVOTO) or insufficient pulmonary valve function. These problems are typically a part of the Tetralogy of Fallot (TOF), which is the most common CHD with an estimated prevalence of 0.26 to 0.48 per 1,000 live births (1, 2). A visual comparison of the healthy heart and the heart suffering from TOF is provided in Figure 1.
Other CHD (e.g. pulmonary atresia, regurgitation, or stenosis) that warrant PVR are less common. In case of congenital aortic valve stenosis, the Ross procedure is used to remove the pulmonary valve to serve as an aortic valve (AV), so the pulmonary valve has to be replaced (1). Later in life, the dysfunctional RV to pulmonary connection occurs most often because of valve degeneration after replacement. Infections, cancer, cardiovascular risk factors, and degeneration can be other reasons (3).
Figure 1. Normal heart vs. TOF heart. Reproduced with courtesy of WHO (4).

PVR is a successful procedure because it provides a well-functioning PV, reduces QRS duration, and improves RV volume and function (5). Resolving those issues improves the patients’ prognoses because dilated RVs with poor functionality and prolonged QRS duration (> 180 ms) were associated with ventricular tachycardia, syncope, and sudden death in TOF patients (1, 6). PVR is a safe procedure with a low mortality risk (< 1%) and high freedom from reoperation after longer time periods (7-9), and current guidelines for assessing PVR operation and reoperation indications provide specific evidence-based recommendations. However, the current number of 290,000 heart valve replacement procedures is expected to triple by the year 2050 (10). That is why further improvements are necessary to increase the durability and functional reliability of the existing valves (i.e. xenograft, mechanical, homograft, monocusp, bicuspid, tissue-engineered etc.).

1.1. PVR Indications
One of the key indications for PVR reoperation in patients with TOF or pulmonary stenosis is pulmonary regurgitation (PR). PR is the backward blood flow through the PV into the RV that is determined by the difference in diastolic pressure between the pulmonary artery and the RV (11). This can be seen early after operation, being attached to a positive pressure ventilator or in a chronic setting, when RV stiffness lowers and RV afterload elevates. Although severe PR is a reliable indication for TOF reoperation, it is difficult to evaluate RV diastolic function to assess RV stiffness. Furthermore, it is suggested that RV stiffness in patients with repaired TOF could be beneficial because it prevents RV dilation and maintains a suitable level of exercise capacity (6). Therefore, even when patients are suffering from the loading effect of chronic PR, RV dilatation may not occur when RV stiffness is high.

In addition to screening patients for PR, it is important to monitor progressive RV dilatation and progressive decreases in RV functions to ensure that PVR reoperation is warranted. RV dilation in patients after TOF repair occurs because of the increased RV preload due to PR, but it has also been correlated with other issues, such as outflow tract aneurysms (11). Many causes have been associated with RV functional failure, but RV systolic dysfunction and tricuspid regurgitation (TR) are the main PVR indications for symptomatic TOF patients (7, 12).
The European Society of Cardiology (ESC) recommends PVR in symptomatic TOF patients with severe PR and/or stenosis with following diagnostic evidence: RV systolic pressure > 60 mmHg and TR velocity > 3.5 m/s (7). PVR is also recommended in asymptomatic TOF patients when they fulfill at least one of the following six criteria: (a) measured decrease in exercise capacity, (b) progressive RV volume dilation, (c) progressive RV systolic dysfunction, (d) progressive TR categorized as moderate, (e) RV outflow tract obstruction (RV systolic pressure > 80 mmHg and TR velocity > 4.3 m/s), and (f) consistent atrial or ventricular arrhythmias (7). In both cases, evidence level C is used to provide guidelines, which means that the evidence is based on expert opinions and less significant studies (i.e. small sample sizes, registries, or retrospective studies). The American College of Cardiology (ACC) and the American Heart Association (AHA) have similar recommendations for PVR, but their recommendations are supported by evidence level B (i.e. basing recommendations on large non-randomized studies or one randomized controlled trials) (9). However, both guidelines agree that the main indications for PVR are RV size and function (7, 9). It is suggested that end-diastolic volume (EDV) and end-systolic volume (ESV) are indications for PVR when EDV > 150 mL/m2 and ESV > 80 mL/m2 (13). A slightly higher RV EDV (> 163 mL/m2) was reported as a PVR indication by Lee et al. (14). Other indications for PVR include EF (RV < 47%; LV < 55%), QRS duration (> 140 ms), and RV/LV EDV ratio (> 2) (13).
Although RV volumes and functions are currently considered the primary PVR indications, it has been suggested that the importance of LV dysfunctions in patients after TOF repair should not be underestimated (11). Current findings suggest that RV EF and LV EF are correlated, so LV dysfunctions are a potential indicator for RV dysfunction (13). Changes in the LV occur because the RV pressure- or volume-overload leads to RV dilatation, consequently changing the geometry of the LV and increasing the LV end-diastolic pressure (15). However, it is important to note that LV deterioration in patients after TOF repair is caused by RV dysfunction, and it has been suggested that there is a time gap between RV deterioration and LV deterioration because LV functions are preserved after the initial pathological changes in the RV (13). Therefore, reoperations in TOF patients should aim primarily to assess the RV volume and functions instead of relying on LV dysfunctions because that would probably result in postponing an intervention unnecessarily.

1.1.1. MRI in patient evaluation
The primary diagnostic test used to measure these indications for PVR is magnetic resonance imaging (MRI), and several MRI techniques for screening RV volume and functions have been tested, validated, and recommended for clinical practice using several different procedures (16). Sheehan et al. (17) found that using a knowledge-based reconstruction model in MRI shows slight bias and overestimates ESV, EDV, and EF, but the error was not considered statistically significant. Mooij et al. (18) tested the ECG-gated steady-state free precession imaging sequences technique for measuring RV mass, volume, and function using the used standard protocols to assess morphology and function, and they concluded that MRI assessments of RV measurements are reproducible and reliable. A comparison of axial and short-axis orientations showed that both methods can be used to obtain reproducible measurements of RV EDV, ESV, EF, and stroke volumes (19). Significant differences in reproducibility between cine and spine-echo MRI were not reported in previous MRI evaluations (20).
Regardless of the techniques and orientations used, the MRI can be used to provide excellent images of the RVOT, aorta, pulmonary arteries, and aortopulmonary collaterals. It can also accurately quantify several key measurements for determining PVR, including biventricular size, myocardial viability, TR, PR, and cardiac output (16). Examples of MRI images that warrant PVR after TOF repair because of the large RV outflow aneurysm are shown in Figure 2.
Figure 2. A cine SSFP image (A), oblique sagittal image (B), and an axial plane image (C) that show a thin-walled aneurysm (An) and the associated thrombus (arrow). Reproduced with courtesy of Villafane et al. (16).

1.1.2. Echocardiography in patient evaluation
Echocardiography is a suitable alternative to MRI, but the use of MRI as the primary assessment tool for PVR indications is preferred because each patient can have different indications for PVR, such as pulmonary regurgitation (PR) alone or primarily RVOTO (21). Unlike echocardiography, the MRI can perform various measurements accurately with high intra- and inter-observer variability in addition to quantifying volumes, stroke volumes, mass, and EF for the LV and RV. Other quantified measurements that can be obtained by the MRI include quantified PR, TR, pulmonary-to-systemic flow ratio, and cardiac output, but the images obtained also allow the evaluation of abnormalities in cardiac wall motional, right ventricular outflow tract (RVOT) anatomy, aortic abnormalities, myocardial viability, residual intra- and extra-cardiac shunt, AV regurgitation, aortic size, and coronary arteries (16).

While echocardiography is an excellent tool to describe obstructions with the finding of high velocities in Doppler measurements of the blood flow, it is not comparable to MRI when it comes to PR or RV function assessment (21, 22). False-positive or false-negative outcomes are more likely with echocardiography than with MRI. However, because MRI is a time-consuming and expensive test, various researchers attempted to create methodologies to use echocardiography for assessing PVR indications in patients, but the results are mostly in favor of the MRI (23). Traditional methods for assessing patients using echocardiography cannot overcome the challenge of assessing RV volume and functionality because of the complex morphological shape of the RV, which is something that does not affect MRI assessments (16), so various methods were developed to address that limitation.

1.1.3. New echocardiography methods for RV assessment
Some of the tested echocardiography techniques for assessing the RV are the three-dimensional (3D) volume acquisition analysis, the four-chamber (4C) area, and the 3D knowledge-based reconstruction (3D KBR).

The 3D echocardiography method acquires 3D volumetric data sets from apical 4C views at the rate of 20 – 30 frames per second (24). That method was tested and validated using in vitro studies, animal model studies, and studies with the healthy pediatric population (25), but comparisons with MRI in patients with CHD show contradictory results. Crean et al. (26) found that three-dimensional (3D) echocardiography underestimates RV EDV (-34%), RV ESV (-42%), and overestimates EV EF (13%) when compared to MRI. Dragulescu et al. (24) also found that 3D echocardiography significantly underestimates EDV (7%, p < 0.001) values compared to MRI, and it also had the highest inter-observer variability compared to other echocardiography methods (EDV mean bias = -13.4 ± 17.6, p = 0.01; ESV mean bias = -11.6 ± 16.5, p = 0.008). Given the poor performance and high variability, the 3D model was not recommended for clinical practice.

The 4C method measures the RV end-diastolic area from the apical 4C view, which is then indexed to body surface area to estimate the EDV index (24). Alghamdi et al. (22) tested the 4C method and estimated that it reduces the need for MRI in 29 – 41% of the patients, but one false-negative result could be expected when using 4C echocardiography. Dragulescu et al. (24) found that inter-observer differences for the 4C-derived EDV values are significant (-33 ± 29.1, p = 0.0001), and it also overestimated RV volumes by 5.9% with a wide range of differences between reported values (SD = 33.1 mL). The intra- and inter-observer coefficients of variation reported by Dragulescu et al. (24) are 8.5 and 10.8 while the coefficients of variation reported by Alghamdi et al. (22) are 14.7 and 9.6 respectively.

The 3D KBR method (VentriPoint Medical System [VMS], Ventripoint

Diagnostics Ltd., Seattle, WA, USA) has been described by Kutty et al. (27) and Dragulescu et al. (24), 3D KBR obtains two-dimensional (2D) echocardiographic images while using a magnetic tracking system (see Figure 3, top panel). The anatomical structure points on those 2D images are identified in the magnetic 3D space (see Figure 3, middle panel), followed by a full 3D surface reconstruction of the RV (see Figure 3, bottom panel). The 3D surface reconstruction is performed by using a database that contains the parameters of RV shapes of anatomically similar patients. That is how ESV, EDV, and EF values for the RV are calculated. Compared to 4C and 3D echography, the 3D KBR method is considered the most accurate assessment because of low deviation from MRI-derived values (24). Kutty et al. (27) found 3D KBR is comparable to MRI for assessing RV volumes and functions in adolescents and young adults with a d- transportation of great arteries repaired using an atrial switch. Dragulescu et al. (24, 28) tested the 3D KBR method in young patients, mean ages of 14.2 ± 7.3 years and 13.7 ± 2.8 years respectively, after TOF repair, and concluded that this method is highly feasible and reproducible in that group. This method has not been tested in the adult population, but current results show promise for future research and possible implications in clinical practice.
Figure 3. Systemic RV reconstruction using 3D-KBR Reproduced with courtesy of Kutty et al. (27).

1.1.4. Exercise testing
The role of exercise testing for monitoring indications for PVR has also been discussed because the maximum oxygen uptake and the minute ventilation/production of CO2 slope correlates with survival rates and quality of life in heart failure patients and has been associated with PVR patients (29). Geva (13) suggests that exercise testing can be a feasible indication for PVR in symptomatic patients after TOF repair when the decreased maximum oxygen uptake (< 70% of reference standard for age and gender) that cannot be explained by chronotropic incompetence. Exercise intolerance has been associated with higher risk for reoperation in patients after TOF repair, but it cannot be used to predict all hemodynamic indications. For example, it was reported that exercise capacity does not reduce in patients with TR (30), but it has been associated with chronic PR (31, 32). Babu-Narayan et al. (33) recommended preoperative cardiopulmonary exercise testing because low peak oxygen uptake was the strongest predictor of high early mortality risk (OR = 0.65 per 1 mL/kg/min, p = 0.041).

1.2. PVR Types
The most common types of valves used in PVR in our center include homografts, biological valves, and monocusps because those methods show the lowest risk for reoperation (34). However, the exact procedures used to repair the RV to pulmonary connection depend on the age of the patient and the anatomical findings. For example, it is difficult to make anatomical repairs in infants and children, so the use of shunts or percutaneously delivered stents is an option to ensure blood flow reaches to the lungs (35). Percutaneous delivery of stents can also be used later in life. The stents used in patients above a body weight of 20 kg and above are valve bearing and show only mild PR at short-term follow-up (36, 37). The procedure has matured and nowadays an anchor stent is often placed before the valve bearing stent is used. That reduces stent fractures and makes it possible to place in already stented valves as well as in homografts. However, certain conditions like RVOT dilatations still have to be addressed surgically. Coats et al. (37) reported that the freedom from reoperation at 1 year in patients who received percutaneous interventions to the RVOT was lower than in patient who received a surgical intervention (86.1% vs. 100%). Although percutaneously delivered valve bearing stents do not show the same level of effectiveness as surgical procedures, they are preferred in cases when non-invasive procedures are possible (38). The Melody® valve and Sapiens™ valve are commercially available (39). They use a bovine venous valve and a bovine pericardial valve, respectively. They have to therefore be considered biological valves and their performance is expected to be similar. Lurz et al. (38) reported freedom from reoperation at 84 ± 4% and 70 ± 13% and freedom from transcatheter reintervention at 73 ± 6% and 73 ± 6%, at 50 and 70 months, respectively. Overall survival at 83 months was 96.9% (38).

Using transanular patches in infant procedures is very common in infants because it widens a too narrow outflow tract (40). The use of transanular patches has also been reported in adult patients undergoing complete TOF repair for the first time with good long-term survival (72.8 ± 17.1% at 14 years) and significant functional improvements according to the NYHA functional class (41). The monocusp valve is often used in combination with transanular patches by placing an additional, flexible patch under the transanular patch, so backwards flow is prevented while forward flow occurs when the systolic blood flow presses the patch to the wall (42). A follow-up on patients with a polytetrafluoroethylene monocusp valve showed a rate of freedom from reoperation of 86%at 1 year, 68% at 5 years, and 48% at 10 years (43). Bicuspid valves were tested because superior durability in comparison with monocusp valves was reported, and they are also suitable for use in children while maintaining their function as the native tissue grows (44). Polytetrafluoroethylene bicuspid valves have been tested in both children and adults, and a short follow-up (1.5 years) from 2005 found significant improvement in the NYHA classification, RV end diastolic area, and pulmonary insufficiency (45). Studies with longer follow-up periods are warranted to confirm their effectiveness.

1.2.1. Homografts
A homograft is a pulmonary or aortic valve from a human cadaver (allograft) or the patient (autograft). The first homograft procedures were performed in 1962 by Ross and Barrat-Boyes, and the procedure at that time involved aseptically harvesting the homograft, sterilizing it, storing it in a nutritional solution, and implanting it within a few days (46, 47). Sterilization was conducted using ethylene oxide, irradiation, or betapropiolactone while the storage of the homograft included placing it in Hank’s solution (at 4 °C) or freeze drying for preservation purposes (46). However, these sterilization procedures were discontinued in 1968 when antibiotic sterilization was introduced, and the development of cryopreservation in 1975 increased the possible homograft storage length from 1 month to 10 years (46, 47). The development of cryopreservation was also significant because cryopreserved homografts still contain viable cells while freeze-dried homografts are considered nonviable as they lose cellular viability (48). The use of detergent decellularization in contemporary homograft processing enables the removal of more than 99% of cells and cell debris while preserving the extracellular matrix (49). Therefore, the main contemporary homograft types used today include the cryopreserved homograft and the fresh decellularized pulmonary homograft (DPH). Depending on their origin, homografts are divided into pulmonary and aortic homografts.

Overall, homografts are a reliable and durable replacement valve. The study by Matsuki et al. (50) found that the homograft is a reliable replacement valve when used for the purpose of RVOT reconstruction while using a homograft for aortic valve replacement has poor long-term durability and low reoperation-free survival rates, but that study reported using outdated preservation methods that affect the cellular viability of the homografts. Later studies documented higher survival rates for patients with homograft valves in the aortic position (46, 47), which could be explain by the transition from betapropiolactone-treated homografts preserved in nutritional solutions to cryopreserved homografts. However, the overall long-term mortality and low improvements in New York Heart Association (NYHA) class suggest that the application of homograft valves in aortic positions is not suitable (46). However, the effectiveness and safety of homograft valves in PVR is supported by evidence. All studies that reviewed survival rates higher than 84.5% for the mid-term period (46, 47, 49, 51) while a long-term follow-up of 20 years showed a 77.8% survival rate (50). Detailed results from follow-up trials involving homograft valves are reported in Table 1.

Although homografts can now be decellularized and cryopreserved, limited availability is still a significant limitation of homograft valves (56). In order to overcome that limitation, mechanical and xenograft (i.e. biological valves from other species) valves have been developed.

1.2.2. Mechanical valves
The first heart valve replacement was performed by Hufnagel in 1952 by inserting the valve in the descending aorta (10). Some of the early valve designs were the ball-and-cage valves, caged-disc valves, and tilting-disc valves, but most current designs are based on the bileaflet design (St. Jude Medical Inc., Minneapolis, MN, USA) because of its superior biocompatibility, thromboresistance, and durability compared to other valves (10).
Although the high-profile configuration and excessive turbulence associated with the first mechanical valves were addressed by contemporary valves, mechanical valves are rarely used because of several limitations. According to Horer et al. (57), mechanical valves are superior to the homograft in terms of hemodynamic performance, but they have no significant difference in reoperation rates when compared to homografts. Notzgold et al. (58) found that patients with autograft valves reported higher perceived quality of life than patients with mechanical valves, and they also have a better NYHA functional class. Mechanical valves are not recommended for children because of their small size and life-long risk for thromboembolism, but they can be used in patients who are already receiving life-long anticoagulant therapy for other medical reasons (59). The ACC/AHA guidelines recommend the use of mechanical valves only in patients who are already taking anticoagulation therapy with warfarin for other medical conditions (9). Mechanical valves are considered to be safe as long as they are chosen correctly by the surgeon while the patient needs to adhere to the anticoagulation therapy, endocarditis prophylaxis, and follow-up (57, 60). The rate of serious prosthetic heart valve complications is approximately 3% per year. They include bleeding, systemic embolization, obstruction due to thrombus or pannus formation, patient-prosthesis mismatch, infective endocarditis, structural deterioration, prosthetic and peri-prosthetic regurgitation, and hemolysis (61).

The use of mechanical valves is also recommended for patients who need PVR late after TOF repair, and they are generally considered a safe alternative to biological valves as long as the patients are provided with an adequate anticoagulation therapy (62-64). The study by Deorsola et al. (65), which focused on four patients at 16, 17, 17, and 18 years of age with 3 or more previous cardiac surgeries, found that there were no adverse events and no need for further reoperations. The study by Ovcina et al. (66) focused on patients who had already undergone PVR with a homograft or xenograft valve and required a valve replacement, and after receiving a mechanical valve, they did not require further reoperations over the next 5 years. Therefore, the use of mechanical valves could be justified in patients with late TOF repair or multiple previous cardiac operations.

However, the majority of studies identified in the literature search that assessed mechanical valves have a small sample size ranging from 4 to 54 participants (see Table 2). Therefore, the generalizability of those findings is questionable. Furthermore, only the study by Stulak et al. (64) found that mechanical valves have higher freedom from reoperation at 10 years than bioprosthetic valves. Other studies reported similar freedom from reoperation when compared to bioprosthetic and homograft valves (57, 60). Given the contradictory findings, there is no conclusive evidence that mechanical valves are safer in terms of freedom from reoperation when compared to other valves, but they are the only type of valve that requires consistent anticoagulation to prevent complications, which is why the development of more convenient replacement valves needs to be pursued.

The current body of knowledge in PVR suggests that biological valves are a reliable solution. Unlike mechanical valves, biological valves do not require anticoagulation therapy, and they can always be available (10). Biological valves are readily commercially available from different manufacturers, but there is no consensus which biological valve is performing the best (56). Even though non-stented valves show faster ventricular regression within 1 year after surgery than stented valves (69), stented bovine pericardium valves are preferred to non-stented ones because of increased durability (70). Although porcine valves are still in use today (e.g. Carpentier-Edwards Bioprosthesis), it is suggested that Carpentier-Edwards bovine pericardial valves, which were introduced in 1981, have superior durability and hemodynamic performance compared to previous bovine pericardial and porcine valves (10). As the Perimount bovine pericardium valve is in focus of this thesis, it will be discussed exclusively (see Table 3).

The Carpentier-Edwards Perimount® valve (Edwards Lifesciences LLC, Irvine, CA, USA) is a stented bovine pericardial prosthesis that is glutaraldehyde-fixed under low pressure and anti-calcification treated in order to maintain cusp compliance (74). The mechanism of calcification in biopresthetic valves is probably related to an inability of the non-viable cells to maintain their normally low intra-cellular concentration of calcium (3). As a result, mineralization occurs when calcium binds to non-viable organelles, collagen, elastin, and other interstitial cellular debris. In addition to the above, unbound glutaraldehyde or its polymers used in the fixation process may become vulnerable to calcification. The Perimount valves are treated with the surfactant polysorbate 80 and ethanol (XenoLogix treatment; Edwards Lifesciences LLC) to extract the phospholipids from the pericardial tissue, change the collagen structure, and controll residual aldehydes. Although that process reduces calcification remains the main cause of failure of the Perimount pericardial valve, and is a time-dependent process (74).

There is no conclusive evidence that Perimount valves are superior to other types of valves, but they show good reliability and durability in all studies. A 1 year follow-up on patients with Sorin and Perimount replacement valves did not show significant short-term differences in survival or freedom from reoperation rates between the groups (69). A comparison of the Perimount valve and other biological valves (i.e. homograft and Medtronic mosaic porcine valves) found similar valve performance among all groups (73). Jang et al. (29) reported that the Perimount valve had the lowest freedom from reoperation at 7 years compared to the Hancock II valve (74.7 ± 7.4% vs. 97.5 ± 2.5%, p = 0.324) and the Biocor valve (74.7 ± 7.4% vs. 100 ± 0.0%, p = 0.368), but the differences were not statistically significant. A comparison of the Perimount valve and the Hancock II valve showed no significant difference between freedom from reoperation for either group (p = 0.51) because the short follow-up duration for the Hancock II group could be the reason for the lack of reoperations (74). A retrospective chart review showed that Perimount valves are comparable to other replacement valves, including the Contegra® conduit, homografts, Shelhigh, porcine xenografts, and porcine-valved conduits at 2 years after surgery (72). Aupart et al. (80) and Marchand et al. (81) reported good performance of the Perimount valve in both the aortic and mitral positions. Bowater et al. (76) reported insignificant PR (< moderate), good durability, and low risk of late valve dysfunction in adult patients.

Two long-term follow-up studies have been conducted on Perimount valve in clinical trials. Aupart et al. (80) reported low rates of valve-related events at 18 years, and the number of patients free from embolism (92 ± 2%), endocarditis (93 ± 4%), hemorrhage (95 ± 2%), reoperation (62 ± 11%), valve failure (68 ± 12%), and other complications (47 ± 8%) was satisfactory for such a long follow-up. Marchand et al. (81) also reported low thromboembolism (83.8% ± 3.2%), hemorrhage (86.6% ± 3.2%), and structural valve deterioration (68.8% ± 4.7%) at 14 years.

1.2.4. New PVR technologies
Despite the success of current biological valves, tissue-engineered pulmonary valves (TEPVs) are being developed to further increase PVR safety, durability, and functionality. The current biological valves display two main weaknesses. First, even though decellularization is currently used to treat homografts before PVR, the patients’ immunological responses are still possible. Second, current biological valves have a fixed diameter that does not adapt to native tissue growth, which significantly reduces the durability of existing replacement valves, especially in infants and children. In theory, it is expected that TEPVs would overcome these limitations because of the repopulation of the valve material with appropriate cells (82). The possibility of creating valves from tissues analogous to the native heart valve tissues is specific to TEPVs, in which case they would show superior durability and fewer side effects (83). That is an important advantage of TEPVs over other valves and a valid reason for pursuing their development because it is currently estimated that 80% of patients with bioprosthetic valves will require a reoperation or display valve dysfunction within 10 years after surgery (70). Geva (84) considers developing a suitable bioengineered pulmonary replacement valves is necessary in order to provide CHD patients with long-term solutions because current replacement valves and procedures provide palliation rather than a solution.

Some examples of methods used to create TEPVs include repopulating biological valves with appropriate cell types (both in vitro and in situ), assembling biodegradable valve matrices, and various other methods. Various autologous and allogenic cells, such as stem or progenitor cells, are being tested in populating different allogenic and xenogenic tissues in order to reduce immunological responses to biological replacement valves. Both synthetic (e.g. polyglycolic acid) and biological materials (e.g. collagen, alginate) are used to create TEPVs (82). However, each material has certain disadvantages. For example, collagen is difficult to obtain and may cause an immunological response or inflammation while autologous fibrin does not initiate an immunological response, but it lacks the mechanical properties of collagen and synthetic materials (10). Currently, the CorMatrix® patch, which is made from decellularized porcine small intestinal submucosa extracellular matrix, is one of the most promising materials in tissue-engineering because of its ease of use and high availability. The growth and durability of the CorMatrix® patch have been speculated based on current research, but longitudinal clinical trials are needed to support those speculations (85). Ultimately, the goal is to create an autologous prosthesis with regenerative capabilities, which would provide a permanent solution to CHD patients that need PVR (86).

Most of the current literature on TEPVs is based on in vitro studies and animal models in order to test valves before they are used in human trials. Some notable systems for evaluating TEPV in vitro performance include the Pearson system that stimulates in vivo conditions, the Goldstein system that replicates the dynamic flow environment for testing aortic replacement valves, Dumont system for mimicking the circulation system, and the Hoerstrup pulse duplicator that gradually increases flow and pressure (10). The hemodynamic functions of bone marrow-derived pulmonary replacement valves have been established in vivo (87), and they are currently being used in animal models. Metzner et al. (88) populated pulmonary valved stents with carotid artery-derived endothelial cells and smooth muscle cells (group 1; n = 5) and bone marrow-derived auotlogous CD133+- cells (group 2; n = 5). Group 2 showed no inflammation or calcification while group 1 showed mild levels of inflammation and calcification. The 3-week follow-up study found that slight calcification occurred in group 2, but there were no signs of inflammation in group 2, which had a significantly lower transvalvular gradient than group 1 (89). The early in vivo results with stem cells and human fibroblasts show promising results (90, 91), but further optimization of static and dynamic mechanical properties of these valves in animal models is needed before they can be tested on human participants.

Some studies with TEPVs have already been conducted on human participants with excellent results. One study with eleven patient, mean age 39.6 ± 10.3 years, that received decellularized homografts with seeded autologous endothelial cells reported no signs of calcification in up to 10 years (92). CorMatrix® patches have been tested in CHD patients, but the behavior of the material was not consistent with studies using animal models because the patients experienced a strong inflammatory response while the patches remained acellular instead of remodeling according to the native tissue (93). However, the case study with a 12-year-old female patient by Gilbert et al. (94) showed promising results within 5 months of follow-up when a trileaflet pulmonary valved conduit was constructed of CorMatrix® patches.
It is considered that the variability of reported outcomes could be attributed to the fact that transition from animal models to human trials is relatively new in TEPV studies, so it is possible that the performance of CorMatrix® patches will improve over time.

Surgical procedures used to deliver replacement valves are also advancing. A cohort of 220 patients that showed 92% survival rate in patients operated before 2005 and a 98% survival rate in patients operated between 2005 and 2010, p = 0.019 (33). Antenatal corrective cardiac surgery and robotically-assisted cardiac surgery are some of the new surgical procedures in early stages of research that could further improve CHD repair procedures (85). Pre-clinical data suggests that the use of TEPVs in transcatheter valve replacement procedures will possibly reduce the degeneration rates associated with bioprosthetics (95). For example, decellularized scaffolds and polymer scaffolds show promising results for RVOT reconstruction. Decellularized scaffolds have already been tested in human trials with no instances of reoperations within 53 months while polymer scaffolds demonstrated in situ histological regeneration properties in the canine model (85). While different procedures and new materials have the potential to revolutionize PVR surgery, it is important to note that cell implantation, growth, and adaption in situ are not yet completely understood, so further work is necessary to clarify those mechanisms and develop appropriate in vitro simulation methodologies for evaluating TEPV performance (83).

2. Main Aims of the Study
The central research questions for these three studies were: (a) “Which type of valve is the best replacement for a pulmonary valve?” and (b) “What are the indications for reoperating a replaced pulmonary valve?” Given the constant development of new valves and assessment methods, two main aims were established. The first purpose was to assess and report the performance of the Perimount biological valve in order to determine its effectiveness compared to other replacement valves. The evaluation of the Perimount valve was considered important in order to compare it to other valves and determine whether it can reduce freedom from reoperation and increase survival rates in patients undergoing PVR. The second purpose was to review and compare well-established and new methods for assessing the indications for reoperation in patients with degenerated artificial valves, and those assessment methods include MRI, traditional echocardiography, treadmill testing, stress echocardiography, and new advanced echocardiography techniques.

The first study was designed as a retrospective overview of PVR procedures carried out at Rikshospitalet with the purpose of providing an overview of different types of valves used and assess the freedom from reoperation associated with each type of valve. The second study was designed as a five year follow-up of patients with Perimount valves identified in the first study because the first study did not provide statistically significant data in terms of follow-up length. These two studies addressed the first aim, which was to evaluate the Perimount valve, and the second study also implemented treadmill testing and stress echocardiography in order to test how measures obtained with these assessment methods would serve as indications for reoperation. The third study evaluated a 3D knowledge-based echocardiography method for assessing PVR indications for reoperation in patients with TOF after pulmonary valve replacement in order to compare it to MRI as the current reference standard. The aim of this study was to test the feasibility of implementing the 3D knowledge-based echocardiography for reoperation risk screening in clinical practice because MRI is a time-consuming and expensive procedure. Therefore, identifying a suitable alternative assessment strategy based on echocardiography could improve the quality of care.

3. Subjects and Methods
3.1. Subjects
The sources used to identify eligible patients included the database “Berte” for the pediatric population and the “Datacor” heart surgeon database, which was used to search for procedures and inputs from perfusionists. Other sources of information included “operation books” to search for surgical procedures documented by surgeons and the electronic patient journal to search for diagnoses and procedures.
The first paper collected medical charts from patients that underwent surgery between January 2000 and December 2007 for primary and late RVOTO repairs that took place in Rikshospitalet, Oslo University Hospital. A total of 286 patient records (n = 67 for under two years of age; n = 143 for 2-9 years of age; n = 76 for 9-17 years of age; n = 79 for over 17 years of age) with 365 surgeries were obtained. Cases involving ventriculo-pulmonary shunts were excluded from the study (34).

The second paper examined 56 patients from a total of 90 patients (mean age 22.8 ± 10.8 years old) that were available 5 years after receiving PVR surgery with Perimount valves (96). The third paper included 30 TOF patients (median age 20 years, age range 8-61) with Carpentier-Edwards Perimount valves (97).

The second and third study obtained approval from the regional ethical committee because the prospective research was carried out on human participants. In compliance with ethical standards in human research, patients were required to provide informed written consent in order to participate in the study. Because children and young adolescents are legally not allowed to provide informed consent on their own, parental written consent was collected when applicable. The first study obtained consent to review hospital charts from Rikshospitalet, Oslo University Hospital, and because of the retrospective nature of the study, an approval from an ethical committee was not required.

3.2. Clinical Examination
The clinical examinations at follow-up in the second and third study included a physical examination, patient anamnesis, blood pressure measurement, bodyweight measurement, height, and ECG testing. The patients participating in these studies were already diagnosed with cardiac conditions prior to their surgical procedures. The chart reviews in the first study revealed that most of the patients underwent PVR because of TOF or pulmonary atresia (n = 185; 65%) while other common reasons for PVR included pulmonary stenosis (n = 27; 9%), common arterial trunk (n = 21; 7%), Ross operated aortic valve disease (n = 20; 7%), transposition of the great arteries (TGA; n = 11; 4%) (34). In the second study, 31 of those patients were diagnosed with TOF (96). The third study included only TOF patients (97).

3.3. Echocardiography
Echocardiography was conducted with the General Electric Vivid 7 machine (GE Healthcare, Milwaukee, WI, USA). The patients were not sedated, and the probe sizes were dependent on body size, and the body surface area was calculated using Mosteller’s formula. Z-scores, which are used to normalize the dimensions of cardiac structures to the patients’ body sizes, were calculated based on Daubeney et al. (98). Both studies performed standard echocardiography, but the third study also acquired two-dimensional transthoracic images to test the new three-dimensional (3D) right ventricular model in order to compare that RV volume and function measurement strategy to MRI (97). The standard echocardiography protocol was used because it is recommended for assessing patients with repaired TOF because it provides comprehensive measurements (99).

3.4. Magnetic Resonance Imaging
A 1.5-Tesla scanner (Avanto; Siemens Medical Systems, Erlangen, Germany) was used to conduct MRI. Both studies conducted the examination without sedating the patients, and the participants’ cardiac function was assessed “using a breath-hold, retrospectively ECG-triggered, segmented, balanced steady-state free precession gradient-echo CINE sequence with minimum echo and repetition times” (96, 97). Ten or more consecutive short-axis images were used to outline the borders in the LV and RV. The third paper reports using long-axis views and right and left outflow tract views in addition to the short- axis views. EDV, EF, ESV, and ventricular mass (third study only) were calculated using the modified Simpson’s method. Trabeculae and papillary muscles were excluded from the analysis because it has been suggested that they significantly alter RV measurements and decrease inter-observer agreement (100).

In the second and third paper examined, not all of the MRI examinations could be undertaken on the same day as echocardiography examinations because of limited resources. The median time between echocardiography and MRI examinations was 3.9 months (0.6 – 11.2 months) for the second study and 32 days (0 – 160 days) for the third study (96, 97). Patients with contraindications, including patients with claustrophobia, pacemakers, metal implants, and spinal deformations, were excluded from taking the MRI. The first paper that is examined here is the only study of the three that did not use MRI testing because it was designed as a retrospective study, so the data was collected from hospital charts (34).

3.5. Surgical Procedures
Most PVR types were consistent in infants (i.e. Contegra, Monocusp, and homograft) while various different valves are used for procedures carried out in adults (34). Homografts, Contegra, and mechanical valves were prevalent in surgical procedures until 2004 while Synergraft was discontinued in 2001 because of poor results (34). The use of bicuspid valves (from 2004) and Perimount valves (from 2005) showed good early results and their use continued in subsequent years (34). The two main types of Perimount valves include “Aortic” valves and “Mitral” valves. Perimount “Mitral” valves are preferred, but Perimount “Aortic” valves are used when small diameters are needed (96).
Mortality is a minor concern in PVR procedures because the surgery outcomes in our center are excellent. Only one unexpected early death was recorded while other deaths are not associated with valve-related incidents and are not directly related to cardiac surgery (96). Overall cardiac mortality related to PVR surgery is estimated at 4% when all causes, including multiorgan failure, pulmonary hypertension, septicaemia, and low cardiac output, are considered (96).

3.6. Exercise Testing
The second paper is the only paper that evaluates valve functions using exercise testing. Stress echocardiography was used to measure tricuspid annular plane systolic excursion (TAPSE), corresponding heart rates, and three velocity measures (i.e. pulmonary valve, tricuspid regurgitation, and aortic valve). The Ergoselect 1200 EL (Ergoline GmbH, Germany) was used to engage participants in cycling in a supine position. The device was set to a 30-degree elevation with a left-sided tilt, and a stepwise protocol with a starting load of 25 W was used to increase the load by 25 W every 2 minutes until the patients reached a 150 bpm heart rate.

The stepwise protocol was chosen in order to protect participants from adverse events. According to the American College of Sports Medicine (ACSM), individuals with known cardiac diagnoses are a high-risk group (101). Therefore, any vigorous exercise test conducted on individuals categorized as high-risk for cardiac complications requires medical supervision and electrocardiographic monitoring (101), which were provided to the participants. Seven participants (12%) were excluded from the stress echocardiography because of poor echocardiography results or other reasons (96). The stepwise protocol provided additional safety to the participants. The vigorousness of the exercise test was controlled, so the exercise would be terminated once the heart rate reached 150 bpm.

Treadmill exercise testing was also used to evaluate performance on a treadmill ergometer (Jaeger Oxycon Delta, Viasys Healthcare GmbH, Germany) by monitoring gas exchange analysis and electrocardiography. Three participants were excluded from the treadmill test because the assessment indicated that they were unable to complete it safely (96). The Oslo protocol was used because it starts with a low incline and speed, which are increased every 4 minutes (i.e. two stages) during the test (102). That approach is similar to other approaches using small increments (e.g. Naughton protocol) and is safe for older individuals and patients with chronic conditions (101). Healthy Norwegian adolescents’ scores from the study by Fredriksen et al. (102) and the ACSM healthy adult reference values (101) were compared with the participants’ results.

3.7. Statistics
The second and third paper used MedCalc™ 12.5 (96, 97); the first paper used “R” and SPSS 16.0 for Windows (34). The second and third paper tested the normality of data distribution using the D’Agostino-Pearson test (103). Paired two-tailed t-tests were used to compare echocardiography results, MRI data, and other continuous variables. The first and second paper also used the Kaplan-Meyer analysis to measure time between valve insertion and valve replacement. Current guidelines for reporting mortality and morbidity after cardiac valve interventions require the use of Kaplan-Meyer or another life table technique because valve-related events need to be reported in a time-related manner (104). The third paper used the Pearson correlation analysis to assess the relationship between the variables, and the Bland-Altman (105) plots were used to illustrate the data. Because the first paper used a retrospective design with only up to two valve replacement surgeries, the mixed effects Cox models analyses was identified as the best model for assessing reoperation hazard. The alpha value for statistical significance was predetermined at 0.05 in all studies.

4. Summary of the Results
4.1. Eight years of pulmonary valve replacement with a suggestion of a promising alternative
This retrospective review of medical charts was conducted in order to assess the surgery results, in which the RV-pulmonary artery connection was replaced or repaired, 2.4 years (range 0 – 8 years) after surgery. The surgeries took place between the year 2000 and the year 2007 at Rikshospitalet, Oslo University Hospital. A total of 365 procedures (n = 286 for primary surgery; n = 79 for reoperations) in 286 patients were identified. The data was analyzed separately for children and adults in order to identify the hazards for reoperation based on age and procedures used.

The comparison of Contegra, Monocusp, and homograft procedures used in infants showed no significant difference in reoperation hazards. The Monocusp in adults proved to be a superior procedure to the homograft in terms of reoperation hazard with a 13% lower reoperation hazard than the homograft (p < 0.03). The homograft was the second most reliable procedure given the fact that reoperations were more common in patients with both Bicusp valves (78% higher reoperation hazard, p < 0.001) and Contegra valves (59% higher reoperation hazard, p < 0.001) valves than the patients with a homograft. Although 100% of the patients with Perimount valves remained reoperation-free, the patients with Perimount valves (n = 90) had the shortest follow-up time, so a valid statistical estimation of reoperation hazard in comparison with other procedures was not possible.
4.2. Results 5 years after pulmonary valve replacement with a bovine pericardial valve

The purpose of this cross-sectional study was to follow-up on 90 patients (n = 80 available, n = 56 agreed to participate) with a biological Perimount valve 5 years after surgery in order to evaluate the functionality and durability of the valve. Exercise testing was used to evaluate the physical capacity of patients with implanted valves, and valve gradient changes and grade of insufficiency were calculated in order to assess valve durability. In order to determine the possible causes of valve degeneration, the study tested the effects of valve size, type of diagnosis, and age on valve degeneration.

The comparison between MRI and echocardiography showed that the MRI test showed significantly lower velocity over the pulmonary valve than echocardiography (1.8, 1.0-3.5 vs. 1.95, 1.2-3.7; p = 0.037). However, the assessment of pulmonary insufficiency in 37 participants showed that echocardiography results correlate with the MRI assessment (r = 0.74, 95% CI 0.55-0-86).
Based on the surgical data prior to the surgeries and the five-year follow-up data, significant differences were observed in the gradient over the pulmonary valve (2.8 ± 1.1 m/s before surgery vs. 2.3 ± 0.7 m/s follow-up; p < 0.001), the grade of insufficiency of the pulmonary valve (2.3 ± 0.9 m/s before surgery vs. 1.1 ± 0.8 m/s follow-up; p < 0.001), oxygen uptake as measured by the treadmill test (65.9 ± 10.0% before surgery vs. 84.9 ± 20.2% follow-up; p = 0.006), and QRS width (137 ± 30 ms early postoperative vs. 144 ± 27 ms follow-up; p = 0.044). Spirometry results showed a capacity of 85 ± 19.5% compared to reference values. The average maximum oxygen uptake measured on the treadmill test was 77 ± 18.7% when compared to reference values.

The gradient and insufficiency of valves were significantly dependent on the participants’ age. When comparing the participants below 15 years of age (n = 29) and above 15 years of age (n = 25), significant differences in valve sizes (25.9 ± 1.7 mm vs. 27.0 ± 1.2 mm; p = 0.007), velocity over the valve (2.6 ± 0.8 m/s vs. 1.8 ± 0.4 m/s; p < 0.001), and grade of insufficiency (1.3 ± 0.8 m/s [n=27] vs. 0.7 ± 0.7 m/s; p = 0.006) are observed. There were no significant differences between groups based on the type Perimount valves, diagnoses, and valve sizes.

4.3. Right ventricular volumes assessed by echocardiographic three‐dimensional knowledge‐based reconstruction compared with magnetic resonance imaging in a clinical setting

This prospective study tested the knowledge-based 3D model that measures RV characteristics using transthoracic echocardiography (VMS version 1.0 with TOF protocol) and compared the results with MRI tests. The purpose of this study was to determine who the knowledge-based 3D model measures up to the current reference standards for RV assessment. Because RV volume and function are the main determinants for reoperating patients with pulmonary valves (7, 9), those measures need to be monitored in patients with pulmonary valves using MRI or echocardiography. MRI is currently considered the reference standard for measuring EDV and ESV as indicators for PVR reoperations (13, 18), but MRI is a time- and resource-consuming, which is why new echocardiography methodologies are being developed to address the MRI limitations. The patients were divided into random subgroups in order to test intra- and inter-observer variability, and the differences between ESV, EDV, and EF mean values were compared.

Intra- and inter-observer differences were not found when comparing MRI-derived mean values while the VMS analysis showed significant interobserver differences for ESV (16.6 mL vs. 16.7 mL, p < 0.01) and EF (4.8% vs. 4.5%, p < 0.01) values. The VMS analysis was compared to the current guidelines (7, 9, 13) for PVR indications. The sensitivity and specificity for ESV > 80 mL/m2 (78% and 86%) and EF < 45% (74% and 43%) were lower when compared to EDV > 150 mL/m2 (100% and 86%), which showed no significant differences between the means. A comparison between the MRI and VMS values showed that EDV and ESV measures were significantly higher when derived from VMS. There were no significant differences between the EF values.

5. General Discussion
The first aim of these studies was to evaluate the performance of the Perimount valve and compare it to other replacement valves. The length of the follow-up in the first study for patients with Perimount valves was too short to allow accurate comparisons with other valves. However, the goal to evaluate the performance of Perimount valves specifically was achieved with a detailed prospective follow-up on patients with Perimount valves from the first study. Therefore, the performance evaluation of Perimount valves is considered thorough and reliable, but the valve can be compared to other valves using only other sources of information, so the first aim was only partially achieved.
The second aim was to evaluate traditional and new methods for screening PVR indications in patients after TOF repair. That aim was achieved in the second and third study. The second study used exercise testing, stress echocardiography, and echocardiography, not only to assess the Perimount valve functions, but also to evaluate the effectiveness of exercise testing for routine screening and compare echocardiography results to MRI-derived measurements. The third study tested the 3D KBR echocardiography method and compared it to MRI using multiple observers in order to measure reproducibility and validity of the 3D KBR as the new method. The results of these studies contribute to the body of knowledge on screening patients after TOF repair and will be used to discuss the future perspectives of new methods in further research and clinical practice.

5.1. Patients
The first study and the 5-year follow-up found that patient age is a significant determinant for reoperation hazard risks, which can probably be attributed to growth. This finding is consistent with the study by Sabate Rotes et al. (55), which found that older age at PVR is predictive of lower reoperation rates (hazard ratio = 0.7, p < 0.001). Weipert et al. (52) also reported that valve size < 15 mm in infants with homograft replacement valves determined the necessity for reoperation within 7 years because of the patients growth. Two other studies on Perimount valve report 36% freedom from reoperation at 10 years (median 16 years of age) and 75% at 7 years (mean 15.5 years of age) that support the finding that valve degeneration and risk for reoperation is significantly associated with younger age (29, 106). However, Aupart et al. (80) and Marchand et al. (81) report significantly better outcomes for adult patients with Perimount valves – more than 83% valve survival at 18 years and 14 years respectively. Bowater et al. (76) also report very high freedom from reoperation rates in adult patients (100% at 5 years; 99.4% at 9 years; 98% at 10 years). The only available solution that claims to accommodate native tissue growth is the hand-sewn bileaflet polytetrafluoroethylene valve (44), but longer follow-up periods are necessary to confirm that assumption. Therefore, the higher valve degeneration in younger patients is expected regardless of the type of replacement valve chosen, and the Perimount valve shows excellent durability in all age groups.

The patients’ diagnoses identified in the first study are consistent with the reported diagnoses in the rest of the PVR literature. TOF and pulmonary atresia were the most common reason for PVR (65%), followed by pulmonary stenosis (9%). Bowater et al. (76) reported that their sample consisted mainly of TOF patients (70%) and pulmonary stenosis patients (20%). Boethig et al. (107) reported TOF (48%), truncus arteriosus (14%), and TGA (11%) as the predominant diagnoses in their sample. A large follow-up of 1,095 patients reported that most patients underwent PVR because of TOF (n = 459) and TGA (n = 232). When compared to other causes, TOF is the most prevalent cause for PVR.

Although the first study did not aim to determine how differences in diagnoses affect freedom from reoperations, the second study found that different diagnoses do not account for valve deterioration and freedom from reoperation. According to the findings by Dearani et al. (108), patients with TGA had the worse freedom from reoperation rate while TOF and pulmonary atresia patients had slightly lower freedom from reoperation than patients with univentricular heart and corrected TGA at 15 years. Higher freedom from degeneration for homografts at 10 years was also reported for the Ross procedure compared to other procedures (107). The findings from the second study about the association between are inconsistent with the current literature, which can probably be explained by the fact that patients with diagnoses other than TOF were underrepresented.

Because CHD is one of the most common causes that warrant PVR, the association between young age and accelerated valve degeneration is a significant challenge for contemporary valves. Infants, children, and adolescents are at higher risk for valve degeneration and reoperation than adults because growth continues until the end of puberty, and current replacement valves are not adaptive and do not support the growth of native tissue. Therefore, most bioprosthetic valves are not suitable for children and adolescents because of the high risk for valve deterioration. Autografts are considered the best choice for children and adolescents as they outperform allografts in long-term follow-ups (59). The goal of tissue-engineering is to develop autograft valves that would adapt to the native tissue and have regenerative characteristics, but until that goal is achieved, improving current replacement valves is necessary to improve survival and reoperation rates in young patients.

5.2. PVR
Various different types of valves were assessed in the first study, including bicuspid, monocusp, contegra, homograft, and Perimount valves. Bicuspid valves were identified as safe, effective, and durable by Quintessenza et al. (45) at 1.5 years (median) of follow-up, but the short follow-up does not allow inferring reliable conclusions about their effectiveness. The study by Nunn et al. (44) was a follow-up study with a maximum follow-up of 5 years, and it concluded that bicuspid valves are safe, effective, and superior to monocusp valves. The bicuspid valve results from the first study are contradictory compared to these findings because they showed acceptable regurgitation, but stenosis was prevalent, so early reoperation was required for 40% of the patients within 4 years of follow-up (34). Perhaps the difference in results could be attributed to the size of the bicuspid valves and procedures used. The first study found that valves with 0.4 mm thick membranes were used while Nunn et al. (44) report using valves with 0.1 mm thick membranes that were hand-sewn on the RVOT. Therefore, it is highly likely that thinner hand-sewn bicuspid valves could prove reliable for reducing reoperation rates in infants and children that are attributed to factors like size and growth at their age, but the evidence is still inconclusive and longer follow-ups are required.

The homograft valves were used in a few infants that displayed pulmonary insufficiency and there was no freedom from reoperation within 4 years of follow-up, but the number of infants with homograft valves was small (n = 6), so there is no conclusive evidence regarding the feasibility of homograft valves in infants. Other studies also often report contradictory findings regarding the causes of homograft reoperation in infants. Baskett et al. (109) suggested that human leukocyte antigen-DR and ABO mismatching were associated with echocardiographic failure, but other studies did not find a statistically significant relationship between ABO mismatch and valve failure (52, 110). The freedom from reoperation rates for patients with irradiated and cryopreserved homografts was significantly lower than for patients with Hancock porcine conduits at 15 years in the study by Dearani et al. (108), but younger age was associated with higher risk for reoperation. Only homografts were used in small children and infants because the Hancock conduit in small children is more likely to cause coronary artery compression, so it is highly likely that age differences could explain why homografts underperformed in that study. All studies on homografts in infants concluded that the underlying diagnoses and young age were predictive of faster valve failure.

The monocusp valves in the first study were 0.1 mm thick and constructed from nonporous polytetraflourethylen. Although their use was justified in cases when homografts were unavailable and in infants because of small cardiac size, pulmonary insufficiency was common early after operation. However, the monocusp group had a high freedom from reoperation (> 70% at 7 years), which indicates a lower risk for reoperation than homografts given the fact that the homograft group consisted of twice as many patients as the monocusp group. Because pulmonary insufficiency was the leading cause of early reoperations, it is highly likely that Gore-Tex monocusp valves implanted in the RVOT could decrease the instances of pulmonary insufficiency and reoperations (42, 43). New materials like CorMatrix™ (CorMatrix Cardiovascular, Inc., Tallahassee, FL, USA) are promising alternatives to polytetraflouretylen that could also accommodate growth. CorMatrix™ uses porcine SIS-ECM, and the feasibility of its use in has been shown in a case study with a promising 5 month follow-up (94).

Although the use of monocusp valves in infants and children was effective, the first study also found that Perimount valves were successfully implanted in children. The youngest patient was 2.6 years old at 13 kg bodyweight. The first study found that freedom from reoperation in 90 patients with Perimount valves was 100% at 2.4 years. This finding is consistent with other studies that found 100% freedom from reoperation at 2.5 years (72, 97), so it is possible to suggest that Perimount valves have excellent short-term reliability.

The follow-up study found that freedom from reoperation was 91% at 5 years. Therefore, these results are consistent with the other studies that reported survival rate and freedom from reoperation at 5 years between 50% and 100% for Perimount patients (74, 76). Compared to other valves from the first study, only homografts demonstrated comparable freedom from reoperation. A similar rate of freedom from reoperation for homografts at 5 years (85%) was also reported by Weipert et al. (52). Fiore et al. (73) and Shinkawa et al. (75) reported 92% and 97.7% freedom from reoperation respectively for patients with Perimount valves at 5 years. Longer follow-up studies on patients with homografts found that freedom from reoperation and survival rates range from 70% to 80% at 15 years (54, 55). Most longer follow-up studies on patients with Perimount valves show high freedom from reoperation from 84% to 98% at 10 years (70, 76, 79), with the exception of Chen et al. (77) who report 36% freedom from reoperation at 10 years. However, Chen et al. (77) reported values for the entire cohort that included patients with both Perimount and Hancock II valves, and compared to other studies, the mean age of their participants was lower, which indicates faster valve degeneration. More long-term studies with follow-up times of more than 10 years are needed to evaluate the Perimount valve so that its durability can be accurately estimated. Based on the current literature, it appears that Perimount bovine pericardial valves and homografts have similar durability, and the limitation of both valves is the inability to provide long-term solutions to young patients suffering from CHD.

Only seven patients from 286 with mechanical valves implanted between 2000 and 2002 were identified in the first study, so they were excluded from the data analysis because of the small sample size. Mechanical valves show good freedom from reoperation between 85.7% and 100% within 10 years after surgery, but the findings by Tokunaga et al. (60) suggest that they are eventually outperformed by bioprosthetic valves in terms of freedom from reoperation (66.7% mechanical vs. 85.7% bioprosthetic at 15 years). However, mechanical valves have been compared to bioprosthetic valves with mixed outcomes (57, 60, 64). According to Fuller et al. (111) successful outcomes and high freedom from reoperation rates are consistent in both mechanical valves and bioprosthetic valves. A significant advantage of bioprosthetics over mechanical valves appears to be the necessity for life-long anticoagulation. Biological replacement valves can also be percutaneously delivered, so they have to be used instead of mechanical valves with inoperable patients (65). That is why the implantation of mechanical valves is usually reserved for patients who are already receiving anticoagulation therapy for other reasons while bioprosthetics are preferred in clinical practice (67).

Although valve degeneration was detected at follow-up (average 2.3 velocity over the valve), it is consistent with other studies that reported mean values of 2.1 – 2.2 velocity over the valve at 1.0 – 3.6 years of follow-up (71, 72, 75, 76). The evaluations of homograft valves by Fiane et al. (53) and Boethig et al. (107) reported freedom from degeneration of 46.6 ± 22% at 7 (for pulmonary valves), 32.3 ± 21.3% at 7 (for aortic valves), 68% at 10 years (for Ross procedure), and 25% at 10 years (for procedures other than Ross). Compared to the 68.8% ± 4.7% structural Perimount valve deterioration rate at 14 years by Marchand et al. (81), the Perimount valve appears to have higher durability than homografts.

In the presented three papers, the studies did not focus on differentiating between operations and percutaneous valve delivery. In the first study, only one procedure was identified in which a transcatheter stented bovine jugular valve Melody was used. In the second study, three patients were excluded due to the fact that they had received a percutaneously delivered heart valve. The field of percutaneously delivered heart valves, including bovine pericardial valves, is growing and will probably show different survival rates and freedom from reoperations once sufficient data is obtained. At the moment, However, based on the findings by Lurz et al. (38) and Coats et al. (37) freedom from reoperation in percutaneously delivered heart valves appears to be lower than freedom from surgical reintervention within 4 years, but there are no significant long-term differences at longer follow-up time. Therefore, it is highly likely that there is no significant difference between surgical and catheter-based PVR is not significant for longer follow-up periods (>4 years) in terms of freedom from reoperation. Percutaneous valve delivery is currently recommended only for high-risk and inoperable patients, but it is expected that complications associated with this procedure will reduce as smaller delivery systems are developed (112).

As pointed out previously, patient age was more likely to affect freedom from reoperation than the type of valve used. Several alternatives to homografts and bovine pericardial valves can be used with children to achieve better long term solutions. A study on 93 children, median age 8 years (1.1 months to 22.4 years), found that freedom from reoperation at 8 years was 100% for porcine valves while freedom from reoperation for homografts was 70% at 8 years (113). Despite the good results for the entire group, Kanter et al. (113) divided the patients based on age and found that freedom from reoperation was 100% at 8 years for children over 3 years of age, but only 39% at 8 years for children under the age of 3, regardless of the type of valve used. RVOT reconstruction using handmade tri-leaflet conduits showed comparable results to homografts, but they did not solve the effect of young age on high rates for reoperation (114). Although some valves were considered superior to homografts in infants, all of these studies suggest that age is a more important determinant for reoperation than the mentioned replacement valves.

However, some valves could address the issues related to valve durability in infants and small children. A promising bioprosthetic valve for RVOT reconstruction in children is the Contegra bovine jugular vein graft. A follow-up study on 133 children, median age 30.9 months (4 days to 3 years), with implanted Contegra bovine jugular grafts found that freedom from reoperation was 80% while survival was at 85.7% at 5 years (115). Sierra et al. (116) reported that Contegra has similar freedom from reoperation rates as blood-group matched homografts (90.7% vs. 93.8% at 7 years) and higher freedom from reoperation rates than the blood-group non-matched homografts (90.7% vs. 66.6% at 7 years). This finding is consistent with the results from the first study, which found that Contegra valves in infants outperformed both homografts and monocusps. While the Perimount valve and homografts show good freedom from reoperation and degeneration in the adult population, Contegra valves could potentially improve PVR in infants because of the material’s soft and pliable characteristics, which are designed to reduce coronary artery compression and similar complications associated with other bioprosthetic valves (108). The early stages of studies on TEPVs in human trials also show promising results for better long-term results in the pediatric population, but human clinical trials are in that area are still new. Until most TEPVs have been sufficiently developed and tested, solutions like the Contegra and CorMatrix™ should be pursued as better alternatives to homografts and bovine pericardial valves in infants and children.

5.3. Indications
At this time, the literature search did not reveal other studies using the VMS system in patients after TOF repair with implanted Perimount, which makes the third study a unique in terms of the targeted population. The median age of 20 years (8-61) also makes the third study the only study on adults that underwent TOF repair as previous studies were carried out in the pediatric population and adolescents (24, 28).
Image quality and personal bias were identified as the main causes for differences in border definitions obtained with 3D KBR echocardiography. The most common and significant occurrence of inter-observer bias was recorded in ESV measurements (16 mL), which consequently resulted in EF estimation biases, but EDV measurements were not subject to inter- or intra-observer bias. Overall, the differences between VMS- and MRI-derived values were small (4.2, 4.6, and 1.9 mL).

The clinical feasibility of several echocardiographic methods, including the 3D volume acquisition analysis and 4C, has already been disproved in several studies (24, 26). However, the 3D KBR technique has been evaluated by various previous researchers that found it to be a reliable method with minimal bias (17, 24, 28). Dragolescu et al. (28) reported a slight underestimation for EDV (2.5%) and ESV (4.6%) in 3D KBR echocardiography when compared to the MRI-derived values. A subsequent study by Dragolescu et al. (24) tested three different echocardiography methods and found that 3D KBR is the most accurate method when compared to MRI measurements with an underestimation of EDV values by 6.6 ± 10.0 (p < 0.001). The study by Niemann et al. (117) showed better consistency between 3D KBR echocardiography and MRI than the third study when the four-dimensional RV analysis method (prototype TomTec software, Unterschleissheim, Germany) was used.

Based on current evidence, MRI is still the standard reference for RV volume and function measurement because of high reproducibility. Mooij et al. (18) reported that inter- and intra-observer reproducibility of RV size and function measurements is high (inter-observer ICC = 0.94 – 0.99; intra-observer ICC = 0.96 – 0.99) while only EF measurements were more difficult to reproduce (ICC = 0.79 – 0.87). Unlike 3D KBR-derived values, MRI-derived values showed no significant inter- or intra-observer differences in the third study. Therefore, these findings are consistent with the current literature and confirm the superiority of MRI to echocardiographic methods in terms of reproducibility. However, compared to other echocardiographic methods aimed at assessing the RV, the 3D KBR echocardiography method for measuring RV EDV, ESV, and EF is currently the most promising method for screening PVR reoperation indications.

The role of exercise testing for monitoring reoperation indications has been addressed in the follow-up study. The maximum oxygen uptake has increased during the follow-up time, but it was still low compared to reference values based on healthy participants. That finding is consistent with most of the current literature. Only the study by Bowater et al. (76) did not find a significant difference between preoperative and postoperative maximum oxygen uptake in patients with Perimount replacement valves (2.0 ± 0.4 ms vs. 2.1 ± 0.4 ms, p = 0.8). Fredriksen et al. (102) found that oxygen uptake increases in adolescents with CHD progressively as it does in healthy adolescents, age range 8-16 years, but it remains below the level achieved by healthy adolescents. The low oxygen uptake in preoperative patients and early after TOF repair is associated with the restrictive RV physiology that reduces cardiac output, but the restrictive RV physiology late after repair diminishes the effects of PR, so exercise tolerance is increased (97).

Previous studies suggested that exercise capacity is inversely related with RV dilation and PR, and a cut-off score for inadequate post-repair maximum oxygen uptake was determined at < 85% as patients with lower values showed higher PR severity (11). Despite the increase of maximum oxygen uptake since before surgery in the second study, the average value was 77 ± 18.7% of predicted based on reference values, which is consistent with the 76.3(67.9%; 92.7)% of predicted oxygen uptake reported by Muller et al. (118). A total of 71 patients in the second study were identified as at risk. It is possible to suggest that both inadequate maximum oxygen uptake and high ESV are responsible for the high rate of reoperation risk in this study group due to the fact that these measures have been inversely correlated. Exercise testing is a simple procedure that can be used often and has fewer restrictions than MRI and 3D KBR echocardiography. If it is inversely correlated with RV volumes, it will be possible to use it more frequently in clinical routine than MRI and echocardiography. It is also a possible alternative for patients with contraindications that prevent them from exposure to magnetic fields.

The current guidelines for PVR state the indications for reoperation clearly, but viewpoints on the future of PVR indications are different. Fuller et al. (111) consider that the development of percutaneous valve will justify the changes to the current criteria for ordering PVR because of the less invasive nature of the procedure, so it is possible that early and aggressive valve replacement will be practiced in the future. The early and aggressive approach to PVR is contradictory to Gaynor’s (119) conservative viewpoint that severe PR and RV volume overload justify PVR only when the patient shows associated symptoms. However, waiting for symptoms, such as signs of heart failure and arrhythmia-induced syncope, to manifest places the patient’s safety at risk (120). Reoperation indications should ideally be identified in time using the current guidelines for asymptomatic patients because the waiting for symptoms to develop may result in irreversible RV and LV dysfunctions (121, 122). It is highly unlikely that a late intervention will cause irreversible damage to the heart structures, but it has been demonstrated that the RV volumes do not normalize in patients who receive PVR with the following preoperative measurements: RV EDV > 170 mL/m and RV ESV > 85 mL/m2 (123). That is why a conservative approach limiting PVR to symptomatic patients is not considered feasible for clinical practice. An earlier and more aggressive approach could be implemented once percutaneous valve delivery is further refined, and an individualized approach could be developed if longitudinal multisite datasets are developed to improve current PVR indications (124). For the time being, surgical PVR procedures need to continue using the current guidelines when monitoring asymptomatic patients to improve survival and long-term freedom from reoperation.

5.4. Follow-up
The functional evaluation of the patients with Perimount valves at 5-years revealed that the patients’ oxygen uptake levels increased while QRS duration, RV volume, and EF were all stable despite valve degeneration. The QRS duration increase did occur over time with an average annual increase of 0.9 ±3.8 ms, but that is a lower value than the previous 1.5 ± 1.2 ms per year previously reported by Gatzoulis et al. (6). The extended QRS duration in patients after RVOT repair is inevitable, but shorter QRS duration is associated with restrictive RV physiology after TOF repair (12). The slower pace of QRS elongation is considered beneficial because it is possibly associated with fewer long-term complications (125). Although the Perimount valve showed higher correlations of insufficiency combined with stenosis compared to previous replacement valves, mild stenosis after PVR has been associated with decreased risks for reoperations (126).

The MRI-derived RV ESV mean value was close to the cut-off point for PVR reoperation (74 ± 16 mL/m2), and there were no significant differences between preoperative and postoperative values for RV ESV (p = 0.222) and RV EDV (p = 0.086). It is possible that the RV volume failed to normalize because of the high preoperative mean value (86 ± 29 mL/m2), which is consistent with the finding that none of the patients with RV ESV > 85 mL/m2 before surgery experienced a normalization of RV volumes (123).

The current ESC follow-up recommendations for patients after TOF repair state that cardiac assessments should be performed at least once per year, and echocardiography should be routinely performed at each visit (127). Although it is stated that all patients should have MRI, the length between follow-up times for MRI assessments is not discussed in the current guidelines (127). The follow-up study found that the assessment of pulmonary insufficiency with echocardiography correlates with MRI (r = 0.74, 95% CI 0.55-0.86), and only the median velocity over the pulmonary replacement valve was significantly lower in MRI. Therefore, this study supports the recommendation that echocardiography could be a routine part of follow-up because the differences between MRI and echocardiography were minimal. A previous follow-up evaluation of serial MRI examinations of RV volumes and functions found that inter- and intra-observer errors are large in those cases because subjective interpretations are highly likely when the MRI is used to assess smaller changes in RV volumes and functions (20). However, echocardiography lacks the reliability for use in ordering PVR, so MRI also needs to be included in follow-up examinations (128). In order to avoid measurement errors, MRI should be used less frequently. It could be used to follow up on poor echocardiographic measurements or confirm positive PVR indications measured by echocardiography.
The length between the follow-up is currently suggested minimally once per year, but this follow-up suggests that a single recommendation is not applicable to all patients. The length of the follow-up time should be significantly different for infants, children, adolescents, and adults because young age was associated with faster valve degradation. Elevations in PR and reductions in exercise capacity are the most common in adolescents between 12 and 13 years of age (102). The infant group with homografts and monocusps from the first study showed high freedom from reoperation at early follow-up. While the number of homografts implanted in infants was small, freedom from reoperation in both valves declined rapidly 3 to 4 years after surgery while Contegra valves started declining within a year after surgery. Based on these findings, infants, young children, and adolescents might require shorter follow-up times, such as every 3 to 6 months, than adult patients.

The shorter follow-up time in children and adolescents is also justified by the fact that their quality of life depends on whether they can lead normal lives along with their peers and perform school-related tasks. The role of exercise testing in clinical practice is not discussed by the ESC guidelines, but it is suggested that high-risk patients for arrhythmia and sudden cardiac death should be restricted to low-intensity activities and avoid isometric exercises while there are no restrictions for patients with stable (127). Therefore, clinical exercise assessments have important implications for children and adolescents with replacement valves because they cannot be expected to perform in physical education based on the same norms as their healthy peers. They should also not be excluded from physical activity completely in order to prevent alienation from their peers, so frequent assessments of their maximum oxygen uptake can be

Follow-up periods of once per year are justified within 5 years after operation in patients with Perimount valves and homografts. Studies with 7- and 10-year follow-up periods still report good valve durability, but the majority of patients will experience some sort of valve-related event within 10 years of surgery as the maximum durability for bioprosthetics is estimated between 10 and 20 years after surgery (86). Therefore, follow-up times are also time-dependent. Current evidence suggests that shorter follow-up periods are warranted for patients that received a replacement valve more than 5 years ago. Shorter follow-up times might also be warranted for patients with mechanical valves. In addition to assessing valve degeneration and PVR reoperation indications, shorter follow-up times may reduce valve-related incidents in patients who lack adherence to anticoagulation therapy or require modifications to their therapy for whatever reason.

5.5. Strength and Limitations
Some limitations specific for each study have to be noted. The first study found that Perimount and homograft are the most reliable procedures in terms of reducing reoperation risks, but the length of follow-up on patients with Perimount valves does not provide adequate statistical power to infer an accurate conclusion. That limitation was addressed in the 5-year follow-up study, which evaluated the durability and reliability of Perimount valves over a longer time period. However, previous researchers suggest that bioprosthetic valve functions remain stable within 5 years after surgery, after which they are likely to decline (70). Therefore, the longer follow-up time shows only valid mid-term results while a longer follow-up time would be necessary to obtain more reliable data. Some data in the first study was inconsistent because of the pre- and post-operative assessment differences and the variety of diagnoses that led to PVR procedures. The follow-up addresses those issues by focusing only on patients with complete data sets.
When MRI and exercise testing were required, some patients had to be excluded from those procedures because of high risks for adverse events. Unlike the third study, the MRI and echocardiography analysis in the follow-up study was not performed by multiple observers. The MRI-derived volumes were obtained using the modified Simpson’s method, and it has been documented that the short-axis method can underestimate RV values (129), so it is highly likely that the data obtained in the second and third study is not an accurate reflection of the true RV values (96, 97). Nevertheless, the use of the modified Simpson’s method is justified given the fact that the piecewise smooth subdivision surface method presented by Hoppe et al. (130) is superior to the standard Simpson’s method (17), and the accurate choice of methodology is considered to be the main reason for excellent intra- and inter-observer agreements in the third study.

Even though the participants’ exercise test scores were compared to reference values of healthy adolescents and adults, the exercise test did not have a control group. Because only patients with Perimount valves were included, the study could not make a comparison of maximum oxygen uptake among groups with different types of replacement valves. Therefore, it is not possible to determine whether Perimount valves are superior to other replacement valves in terms of exercise capacity.

The third study had to exclude some patients because the VMS reconstruction database had missing data for patients with new RVOTs and Perimount valves. Signal loss due to artificial valves occurred in both MRI and echocardiography. The MRI images were obtained in the range of 0 – 160 days after the VMS examination, which could affect the differences in VMS- and MRI-derived values. However, it has been demonstrated that MRI volumes do not show significant changes over a median of 1.8 years (131). It is highly likely that the data obtained in this study is accurate. Unlike previous studies, this study did not use standard views to obtain echocardiographic images, so it is possible that the image quality was less than satisfactory and contributed to the deviations from MRI-derived values.

The echocardiographic assessment used the PR grading system to quantify pulmonary insufficiency, but that approach is semi-quantitative at best. MRI-derived PR values are preferred, and it is suggested that PR expressed as volume is a better determinant of RV preload than PR expressed as the percentage of total outward flow (132). It is possible that the use of MRI-derived PR volume measurements would have shown different results.
A significant limitation in all of these three studies was the single-centered setting, so the external validity of these studies is limited. The data was gathered and analyzed from a single clinical location, so it is highly likely that the results cannot be generalized to other locations because of differences in clinical procedures, differences in staff resources among healthcare institutions, and the differences in critical care standards and regulations among different countries. However, a national overview revealed that most PVR procedures took place in Oslo with the exception of five procedures in Bergen that used homografts. This information was supplied by a surgeon in Bergen via e-mail. Therefore, a better design was not feasible because most of the national PVR data can be obtained in a single location, and a single-centered setting could be generalized at a national level.

The second study had a cross-sectional design, so data was collected at one point in time only. Based on other studies evaluating the performance of Perimount valves, a cross-sectional design is considered feasible because there are no significant changes in freedom from reoperation within 5 years after surgery. However, a case control study would have been better suited for the purpose of evaluating exercise capacity because it would provide multiple data points to explain the behavior of maximum oxygen uptake over time. Case control studies could also be beneficial for longer follow-up studies for two reasons. First, valve degeneration and freedom from reoperation begin to decline faster in follow-up periods longer than 5 years. Second, detecting smaller changes in RV volumes and functions using MRI could increase inter- and intra-observer bias, so frequent assessments in short- and mid-term follow-ups could provide inaccurate data (20). The first study was designed as a retrospective study, but for the purpose of estimating the prevalence of different diagnoses and freedom from reoperation based on types of replacement valves, it is considered a feasible design.

These studies also have several notable strengths. Despite the single-center setting, the wide age range of the participants in all studies improves their external validity. The data was analyzed based on the participants’ age groups, so the results are relevant to both children and adults. With 90 patients in the Perimount group, the first study contained a larger sample size than a lot of other studies on patients with Perimount replacement valves. In comparison, the study by Morales et al. (72) included 26 patients with Perimount; the study by Fiore et al. (73) included 18 patients with Perimount; the study by Kwak et al. (74) included 63 patients with Perimount. The studies by Bowater et al. (76), Aupart et al. (80), and Marchand et al. (81) are the only studies with larger sample sizes than these two studies.

Only 56 of the available patients agreed to participate in the follow-up study. Although the follow-up contained 70% of all available patients, the study did not find significant differences between the cohort and the study group. Most importantly, this study included only participants without missing data, so detailed assessments were available for the data analysis. The MRI-derived values for functional assessment in the follow-up study are comparable to the values presented by the Philadelphia and Toronto group (23, 132).

The third study used a 6-mm slice thickness for MRI, and the short-axis views were aligned to the tricuspid valve and the septum. A previous study that used multislice helical computed tomography reported that slice thickness between 2 and 5 mm showed the lowest overestimation for both LV and RV volume measurements (133). However, typical values reported for slice thickness in MRI studies on patients with TOF repair range between 5 and 8 mm (24, 28, 134), so the chosen slice thickness was consistent with previous literature and protocols for MRI assessments. The alignment used for short-axis views in this study has been previously tested in comparison with the LV short-axis series and axial series, and it proved to be superior to both methods in terms of RV volumetric assessments (134). Therefore, validated MRI methodologies that decrease inter- and intra-observer bias are considered one of the key strengths of this study.

6. Conclusions

6.1. Future Perspectives
Various tissue-engineered materials Dubé et al. (3) demonstrated the feasibility of using human dermal fibroblasts to engineer a living tissue sheet that can be mounted on a stent to produce a heart valve. Other studies reported good results in animal models with bone-marrow-derived CD133+- cells (88, 89). Despite the initial positive outlook of tissue-engineered biomaterials in PVR surgery, they have not been tested in human subjects yet. Before TEPVs can be tested in human trials, they need to be thoroughly tested in animal models to measure their mechanical and biological properties, which include thrombo-resistance, integration with native tissue, deterioration over time, onsite tissue behavior, and various other properties (91). Therefore, it is still too early to predict the potential roles of tissue-engineered biomaterials in PVR. Until sufficient data is obtained from human studies, Perimount biological valves appear to be one of the most reliable biological replacement valves with excellent survival, freedom from reoperation, and functionality.

It is also expected that new RV imaging techniques will be developed to address the challenges of RV assessment in the present. 3D KBR is currently considered the most likely candidate for replacing MRI in routine clinical assessment, and it is likely to replace it as the norm for ordering PVR once the technique is further refined and tested. The development of individualized PVR indication parameters and MRI protocols could further reduce valve-related mortality rates. It will be a long time until molecular imaging techniques that will improve the understanding of the myocardial mechanisms and the accuracy of RV assessments are developed (135). Until then echocardiography should be improved until it can replace MRI in clinical practice.

6.2. Possible Implications for Clinical Practice
It is still too early to propose definitive implications for clinical practice. Overall, the Perimount valve demonstrated a very high survival rate and freedom from reoperation at 2.4 and 5 years in these studies, which is consistent with the majority of other studies on patients with Perimount valves. Although young age appears to facilitate the deterioration of Perimount valves, these valves show high durability and reliability in the adult population. In the pediatric population and adolescents, outgrowing the valves is a major concern, so shorter follow-up times for MRI or echocardiographic screening are required.

Exercise tests are included in existing guidelines only as recommendations for symptomatic patients, in which case decreased oxygen uptake that occurs without another reasonable explanation is considered an indication for valve replacement. However, exercise testing should be routinely used to follow-up on patients after PVR because exercise intolerance is associated with PR and higher risk for reoperation/mortality. The use of exercise testing for assessing patients after TOF repair is a feasible strategy for monitoring RV volume and functioning, but it cannot be used as the only assessment strategy because it does not correlate with TR. More frequent follow-up exercise tests may be required early after PVR and in adolescents between 10 years of age and until the end of puberty. Follow-up tests in adults and late after repair can be conducted annually in combination with MRI testing to evaluate valve gradient, RV and LV volumes, and Doppler velocities.

The use of 3D KBR echocardiography in clinical practice is still not recommended until more studies replicate the findings from the third study. Otherwise, minimal bias is reported when values obtained by 3D KBR are compared to MRI-derived values 3D KBR also has several advantages over MRI, which include ease of use and faster data collection. The weak magnetic field used in VMS testing could open the possibility for monitoring patients that are excluded from MRI assessments to prevent adverse events. Therefore, the implementation of VMS could improve the quality of care delivery in terms of efficiency and accessibility. Echocardiography could eventually replace MRI in clinical practice in many cases. For now, MRI should remain the standard procedure when EF values are needed and for ordering PVR as its reliability has more support than any other echocardiographic method evaluated in the current literature. However, the VMS could be used as a follow-up to monitor indications for reoperation.
Given the fact that the MRI is no longer the only option for assessing PVR indications, patients with replacement valves could also be divided into two separate groups in order to plan assessment options accordingly. One group would be comprised of patients who already fulfill indications by follow-up while the other group would be comprised of patients who are being followed and develop indications later. The rationale for this division is the fact that patients with PVR indications need evaluation methods with more specificity and sensitivity to avoid errors. Most patients in the second group will not require precise measurements using MRI or 3D KBR, so QRS duration and maximum oxygen uptake in this group can be screened routinely and more frequently. MRI or echocardiography can be performed annually in this group, but they can also be ordered in a timely manner based on the changes observed in QRS duration or exercise capacity. As mentioned previously, a conservative approach is not considered suitable for assessing patients with replaced valves because late interventions can lead to irreversible ventricular dysfunctions and functional deterioration. Selecting patients that did not develop PVR indications yet for more frequent and less resource-consuming tests is the first step towards early interventions and preventing adverse outcomes associated with postponing PVR.

6.3. Prospects for Future Research

The 5-year follow-up provided adequate data for assessing Perimount valves, but the length of the follow-up is currently sufficient to provide mid-term results only. A longer follow-up is necessary in order to better assess the durability and risk for reoperation in patients with Perimount valves. The first study did not have sufficient data to compare Perimount valves to other types of valves, such as homografts, monocusps, and bicuspid valves, so the next follow-up could recruit participants with different valve types to allow between-groups comparisons. More studies that compare different types of biological valves are also needed as existing literature is usually inconclusive in that area.

Exercise testing for screening patients after PVR needs further development. Although current guidelines suggest that symptomatic patients will display decreased oxygen uptake, future studies can focus on identifying the effects on the heart utilizing MRI methods of stress testing, providing valuable, comparable data like EF, EDV, ESV, etc. at the same time. Prospective studies are considered necessary to suggest optimization strategies for current PVR indications that will improve the timing for valve replacement in patients at risk (136). Case control studies should be used to further evaluate the role of exercise testing in routine clinical assessments. It is considered that these types of studies will provide valuable data collected in multiple points in time, so that the trajectory of maximum oxygen uptake can be monitored and correlated with RV volumes and functions.
The role of environmental influences, such as patients’ levels of physical activity and nutrition, on variables like maximum oxygen uptake, QRS duration, and RV functions also needs further investigation. These studies would be difficult to construct in terms of validity because the use of self-reported questionnaires would be necessary to evaluate the patients’ activity levels and diet habits. However, they may provide additional insight into the role of patients’ lifestyle decisions on freedom from reoperation and quality of life. Those findings could have important implications in patient education after TOF repair.

The use of thin 0.1 mm polytetrafluorethylen bicuspid or CorMatrix™ valves also appears to be promising for resolving the problem of high reoperation rates at an early age. Thin bicuspid valves or even Monocusps could potentially increase freedom from reoperation because they accommodate the growth and development of the native tissue. However, future studies need to evaluate their effectiveness over longer time periods and investigate which reoperation indications are prevalent in this group of patients.

The findings on freedom from reoperation and valve-related events vary across different studies. However, it is possible to notice that studies with younger patients tend to report lower replacement valve durability than studies on adults. Therefore, it is recommended that future studies on PVR make a clear distinction between the pediatric population and the adult population. Since age is an important determinant for faster valve degeneration, it is possible that some reported outcomes in existing studies are skewed because both children and adults were included in the study.
After the third study, it was speculated that MRI images obtained on the same day as echocardiographic images would show less variability in ESV, EDV, and EF values. Because MRI is a limited resource because of its high demand in both clinical practice and trials, it is not always available, so study designs that use large number of participants will never be able to overcome that limitation. It is possible that studies designed as single or multiple case studies could overcome that limitation. Case studies have low external validity because they use one participant or several participants on a case-by-case basis, but they might be the only feasible research designs for obtaining both echocardiography and MRI measurements on the same day because of the small number of participants. That is why case studies could be considered for future research in this area, at least to provide an estimated effect of time gaps between echocardiography and MRI on deviations in RV measurement values.

Because the VMS-derived EF values were not considered sufficient for clinical practice, future studies should focus on comparing VMS- and MRI-derived EF values. Feasibility in comparison with MRI could depend on the existence of good echocardiographic windows to increase the quality of analysis. Patients with mainly pulmonary stenosis could also be excluded due to the fact that follow-up with Doppler is often sufficient.
Tissue-engineering is considered the future of PVR because TEPVs could address various limitations of homografts and xenografts. However, the few available studies on humans are not always consistent with the results obtained from animal models. As possible limitations of these valves in human models are identified by early studies, they should be resolved and retested consistently. Research on TEPVs should continue focusing on the pediatric population and adolescents because those populations are the most vulnerable to valve deterioration and reoperation, so it is expected that they will benefit the most from self-regenerating valves that adapt to the native tissue characteristics and its growth.

Reference List

1. Allen HD, Driscoll DJ, Shaddy RE, Feltes TF, editors. Moss & Adams heart disease in infants, children, and adolescents: including the fetus and young adult. 8th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2013.
2. Apitz C, Webb GD, Redington AN. Tetralogy of Fallot. Lancet. 2009;374(9699):1462-71.
3. Dubé J, Bourget J, Gauvin R, Lafrance H, Roberge CJ, Auger FA, et al. Progress in developing a living human tissue-engineered tri-leaflet heart valve assembled from tissue produced by the self-assembly approach. Acta Biomater. 2014;10:3563-70.
4. Mendis S, Puska P, Norrving B, editors. Global atlas on cardiovascular disease prevention and control. Geneva, CH: World Health Organization; 2011.
5. Kleinveld G, Joyner RW, Sallee D, Kanter KR, Parks WJ. Hemodynamic and electrocardiographic effects of early pulmonary valve replacement in pediatric patients after transannular complete repair of Tetralogy of Fallot. Pediatr Cardiol. 2006;27(3):329-35.
6. Gatzoulis MA, Balaji S, Webber SA, Siu SC, Hokanson JS, Poile C, et al. Risk factors for arrhythmia and sudden cardiac death late after repair of tetralogy of Fallot: a multicentre study. Lancet. 2000;356(9234):975-81.
7. Baumgartner H, Bonhoeffer P, De Groot NM, de Haan F, Deanfield JE, Galie N, et al. ESC Guidelines for the management of grown-up congenital heart disease (new version 2010). Eur Heart J. 2010;31(23):2915-57.
8. McKenzie ED, Khan MS, Dietzman TW, Guzman-Pruneda FA, Samayoa AX, Liou A, et al. Surgical pulmonary valve replacement: A benchmark for outcomes comparisons. J Thorac Cardiovasc Surg. 2014.
9. Warnes CA, Williams RG, Bashore TM, Child JS, Connolly HM, Dearani JA, et al. ACC/AHA 2008 Guidelines for the Management of Adults with Congenital Heart Disease: Executive Summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (writing committee to develop guidelines for the management of adults with congenital heart disease). Circulation. 2008;118(23):2395-451.
10. Dasi LP, Simon HA, Sucosky P, Yoganathan AP. Fluid mechanics of artificial heart valves. Clin Exp Pharmacol Physiol. 2009;36(2):225-37.
11. Chaturvedi RR, Redington AN. Pulmonary regurgitation in congenital heart disease. Heart. 2007;93(7):880-9.
12. Haddad F, Doyle R, Murphy DJ, Hunt SA. Right ventricular function in cardiovascular disease, part II: pathophysiology, clinical importance, and management of right ventricular failure. Circulation. 2008;117(13):1717-31.
13. Geva T. Repaired tetralogy of Fallot: the roles of cardiovascular magnetic resonance in evaluating pathophysiology and for pulmonary valve replacement decision support. J Cardiovasc Magn Reson. 2011;13:9.
14. Lee C, Kim YM, Lee CH, Kwak JG, Park CS, Song JY, et al. Outcomes of pulmonary valve replacement in 170 patients with chronic pulmonary regurgitation after relief of right ventricular outflow tract obstruction: implications for optimal timing of pulmonary valve replacement. J Am Coll Cardiol. 2012;60(11):1005-14.
15. Haddad F, Hunt SA, Rosenthal DN, Murphy DJ. Right ventricular function in cardiovascular disease, part I: Anatomy, physiology, aging, and functional assessment of the right ventricle. Circulation. 2008;117(11):1436-48.
16. Villafane J, Feinstein JA, Jenkins KJ, Vincent RN, Walsh EP, Dubin AM, et al. Hot topics in tetralogy of Fallot. J Am Coll Cardiol. 2013;62(23):2155-66.
17. Sheehan FH, Kilner PJ, Sahn DJ, Vick GW, 3rd, Stout KK, Ge S, et al. Accuracy of knowledge-based reconstruction for measurement of right ventricular volume and function in patients with tetralogy of Fallot. Am J Cardiol. 2010;105(7):993-9.
18. Mooij CF, de Wit CJ, Graham DA, Powell AJ, Geva T. Reproducibility of MRI measurements of right ventricular size and function in patients with normal and dilated ventricles. J Magn Reson Imaging. 2008;28(1):67-73.
19. Clarke CJ, Gurka MJ, Norton PT, Kramer CM, Hoyer AW. Assessment of the accuracy and reproducibility of RV volume measurements by CMR in congenital heart disease. JACC Cardiovasc Imaging. 2012;5(1):28-37.
20. Pattynama PM, Lamb HJ, Van der Velde EA, Van der Geest RJ, Van der Wall EE, De Roos A. Reproducibility of MRI-derived measurements of right ventricular volumes and myocardial mass. Magn Reson Imaging. 1995;13(1):53-63.
21. Brown DW, McElhinney DB, Araoz PA, Zahn EM, Vincent JA, Cheatham JP, et al. Reliability and accuracy of echocardiographic right heart evaluation in the U.S. Melody Valve Investigational Trial. J Am Soc Echocardiogr. 2012;25(4):383-92 e4.
22. Alghamdi MH, Grosse-Wortmann L, Ahmad N, Mertens L, Friedberg MK. Can simple echocardiographic measures reduce the number of cardiac magnetic resonance imaging studies to diagnose right ventricular enlargement in congenital heart disease? J Am Soc Echocardiogr. 2012;25(5):518-23.
23. Mercer-Rosa L, Yang W, Kutty S, Rychik J, Fogel M, Goldmuntz E. Quantifying Pulmonary Regurgitation and Right Ventricular Function in Surgically Repaired Tetralogy of Fallot: A Comparative Analysis of Echocardiography and Magnetic Resonance Imaging. Circ Cardiovasc Imaging. 2012.
24. Dragulescu A, Grosse-Wortmann L, Fackoury C, Mertens L. Echocardiographic assessment of right ventricular volumes: a comparison of different techniques in children after surgical repair of tetralogy of Fallot. Eur Heart J Cardiovasc Imaging. 2012;13(7):596-604.
25. Seguela PE, Hascoet S, Brierre G, Bongard V, Acar P. Feasibility of three-dimensional transthoracic echocardiography to evaluate right ventricular volumes in children and comparison to left ventricular values. Echocardiogr-J Card. 2012;29(4):492-501.
26. Crean AM, Maredia N, Ballard G, Menezes R, Wharton G, Forster J, et al. 3D Echo systematically underestimates right ventricular volumes compared to cardiovascular magnetic resonance in adult congenital heart disease patients with moderate or severe RV dilatation. J Cardiovasc Magn Reson. 2011;13:78.
27. Kutty S, Li L, Polak A, Gribben P, Danford DA. Echocardiographic knowledge-based reconstruction for quantification of the systemic right ventricle in young adults with repaired D-transposition of great arteries. Am J Cardiol. 2012;109(6):881-8.
28. Dragulescu A, Grosse-Wortmann L, Fackoury C, Riffle S, Waiss M, Jaeggi E, et al. Echocardiographic assessment of right ventricular volumes after surgical repair of tetralogy of Fallot: clinical validation of a new echocardiographic method. J Am Soc Echocardiogr. 2011;24(11):1191-8.
29. Jang W, Kim YJ, Choi K, Lim HG, Kim WH, Lee JR. Mid-term results of bioprosthetic pulmonary valve replacement in pulmonary regurgitation after tetralogy of Fallot repair. Eur J Cardiothorac Surg. 2012;42(1):e1-8.
30. Norgard G, Bjorkhaug A, Vik-Mo H. Effects of impaired lung function and pulmonary regurgitation on maximal exercise capacity in patients with repaired tetralogy of Fallot. Eur Heart J. 1992;13(10):1380-6.
31. Bouzas B, Kilner PJ, Gatzoulis MA. Pulmonary regurgitation: not a benign lesion. Eur Heart J. 2005;26(5):433-9.
32. Gregg D, Foster E. Pulmonary insufficiency is the nexus of late complications in tetralogy of Fallot. Curr Cardiol Rep. 2007;9(4):315-22.
33. Babu-Narayan SV, Diller GP, Gheta RR, Bastin AJ, Karonis T, Li W, et al. Clinical outcomes of surgical pulmonary valve replacement after repair of tetralogy of Fallot and potential prognostic value of preoperative cardiopulmonary exercise testing. Circulation. 2014;129(1):18-27.
34. Neukamm C, Dohlen G, Lindberg HL, Seem E, Norgard G. Eight years of pulmonary valve replacement with a suggestion of a promising alternative. Scand Cardiovasc J. 2011;45(1):41-7.
35. Bonhoeffer P, Boudjemline Y, Saliba Z, Merckx J, Aggoun Y, Bonnet D, et al. Percutaneous replacement of pulmonary valve in a right-ventricle to pulmonary-artery prosthetic conduit with valve dysfunction. Lancet. 2000;356(9239):1403-5.
36. Schreiber C, Horer J, Vogt M, Fratz S, Kunze M, Galm C, et al. A new treatment option for pulmonary valvar insufficiency: first experiences with implantation of a self-expanding stented valve without use of cardiopulmonary bypass. Eur J Cardiothorac Surg. 2007;31(1):26-30.
37. Coats L, Tsang V, Khambadkone S, van Doorn C, Cullen S, Deanfield J, et al. The potential impact of percutaneous pulmonary valve stent implantation on right ventricular outflow tract re-intervention. Eur J Cardiothorac Surg. 2005;27(4):536-43.
38. Lurz P, Bonhoeffer P. Percutaneous implantation of pulmonary valves for treatment of right ventricular outflow tract dysfunction. Cardiol Young. 2008;18(3):260-7.
39. Fleming GA, Hill KD, Green AS, Rhodes JF. Percutaneous pulmonary valve replacement Prog Pediatric Cardiol. 2012;33(2):143-50.
40. Munkhammar P, Cullen S, Jogi P, de Leval M, Elliott M, Norgard G. Early age at repair prevents restrictive right ventricular (RV) physiology after surgery for tetralogy of Fallot (TOF): diastolic RV function after TOF repair in infancy. J Am Coll Cardiol. 1998;32(4):1083-7.
41. Lu X, Wu S, Gu X, Li L, Zhang G, Sun W, et al. Long-term results of surgical treatment of tetralogy of Fallot in adults. Thorac Cardiovasc Surg. 2006;54(5):295-9.
42. Turrentine MW, McCarthy RP, Vijay P, Fiore AC, Brown JW. Polytetrafluoroethylene monocusp valve technique for right ventricular outflow tract reconstruction. Ann Thorac Surg. 2002;74(6):2202-5.
43. Brown JW, Ruzmetov M, Vijay P, Rodefeld MD, Turrentine MW. Right ventricular outflow tract reconstruction with a polytetrafluoroethylene monocusp valve: a twelve-year experience. J Thorac Cardiovasc Surg. 2007;133(5):1336-43.
44. Nunn GR, Bennetts J, Onikul E. Durability of hand-sewn valves in the right ventricular outlet. J Thorac Cardiovasc Surg. 2008;136(2):290-6.
45. Quintessenza JA, Jacobs JP, Morell VO, Giroud JM, Boucek RJ. Initial experience with a bicuspid polytetrafluoroethylene pulmonary valve in 41 children and adults: a new option for right ventricular outflow tract reconstruction. Ann Thorac Surg. 2005;79(3):924-31.
46. Ganguly G, Akhunji ZA, Neethling WM, Hodge AJ. Homograft aortic valve replacement--the experience of one unit. Heart Lung Circ. 2004;13(2):161-7.
47. Gerola LR, Araujo W, Kin HC, Silva GE, Pereira Filho A, Vargas GF, et al. Cryopreserved aortic homograft for aortic valve replacement: immediate results. Arq Bras Cardiol. 2004;83(4):284-7; 0-3.
48. Bodnar E, Matsuki O, Parker R, Ross DN. Viable and nonviable aortic homografts in the subcoronary position: a comparative study. Ann Thorac Surg. 1989;47(6):799-805.
49. Neumann A, Sarikouch S, Breymann T, Cebotari S, Boethig D, Horke A, et al. Early systemic cellular immune response in children and young adults receiving decellularized fresh allografts for pulmonary valve replacement. Tissue Eng Pt A. 2014;20(5-6):1003-11.
50. Matsuki O, Kadoba K, Nakata S, Shirakura R, Nakano S, Matsuda H, et al. Long-term performance of beta-propiolactone-treated nonviable homograft for aortic valve replacement and right ventricular outflow tract reconstruction. Heart Vessels. 1993;8(1):33-6.
51. Cebotari S, Tudorache I, Ciubotaru A, Boethig D, Sarikouch S, Goerler A, et al. Use of fresh decellularized allografts for pulmonary valve replacement may reduce the reoperation rate in children and young adults early report. Circulation. 2011;124(S1):S115-23.
52. Weipert J, Meisner H, Mendler N, Haehnel JC, Homann M, Paek SU, et al. Allograft implantation in pediatric cardiac surgery: surgical experience from 1982 to 1994. Ann Thorac Surg. 1995;60(2 Suppl):S101-4.
53. Fiane AE, Lindberg HL, Seem E, Geiran OR. Homografts for right ventricular outflow tract reconstruction in congenital heart disease. Scand Cardiovasc J. 1997;31(6):351-6.
54. van de Woestijne PC, Mokhles MM, de Jong PL, Witsenburg M, Takkenberg JJM, Bogers AJJC. Right ventricular outflow tract reconstruction with an allograft conduit in patients after tetralogy of Fallot correction: long-term follow-up. Ann Thorac Surg. 2011;92(1):161-6.
55. Sabate Rotes A, Eidem BW, Connolly HM, Bonnichsen CR, Rosedahl JK, Schaff HV, et al. Long-term follow-up after pulmonary valve replacement in repaired tetralogy of fallot. Am J Cardiol. 2014;114(6):901-8.
56. Abbas JR, Hoschtitzky JA. Which is the best tissue valve used in the pulmonary position, late after previous repair of tetralogy of Fallot? Interact Cardiovasc Thorac Surg. 2013;17(5):854-60.
57. Horer J, Vogt M, Stierle U, Cleuziou J, Prodan Z, Schreiber C, et al. A comparative study of mechanical and homograft prostheses in the pulmonary position. Ann Thorac Surg. 2009;88(5):1534-9.
58. Notzold A, Huppe M, Schmidtke C, Blomer P, Uhlig T, Sievers HH. Quality of life in aortic valve replacement: pulmonary autografts versus mechanical prostheses. J Am Coll Cardiol. 2001;37(7):1963-6.
59. Henaine R, Roubertie F, Vergnat M, Ninet J. Valve replacement in children: A challenge for a whole life. Arch Cardiovasc Dis. 2012;105(10):517-28.
60. Tokunaga S, Masuda M, Shiose A, Tomita Y, Morita S, Tominaga R. Isolated pulmonary valve replacement: analysis of 27 years of experience. J Artif Organs. 2008;11(3):130-3.
61. Mankad S. Management of prosthetic heart valve complications. Curr Treat Options Cardiovasc Med. 2012;14(6):608-21.
62. Abbas JR, Hoschtitzky JA. Is there a role for mechanical valve prostheses in pulmonary valve replacement late after tetralogy of Fallot repair? Interact Cardiovasc Thorac Surg. 2014;18(5):661-6.
63. Dos L, Munoz-Guijosa C, Mendez AB, Ginel A, Montiel J, Padro JM, et al. Long term outcome of mechanical valve prosthesis in the pulmonary position. Int J Cardiol. 2011;150(2):173-6.
64. Stulak JM, Dearani JA, Burkhart HM, Connolly HM, Warnes CA, Suri RM, et al. The Increasing Use of Mechanical Pulmonary Valve Replacement Over a 40-Year Period. Ann Thorac Surg. 2010;90(6):2009-15.
65. Deorsola L, Abbruzzese PA, Aidala E, Cascarano MT, S. L, Valori A, et al. Pulmonary valve replacement with mechanical prosthesis: Long-term results in 4 patients. Ann Thorac Surg. 2010;89(6):2036-8.
66. Ovcina I, Knez I, Curcic P, Özkan S, Nagel B, Sorantin E, et al. Pulmonary valve replacement with mechanical prostheses in re-do Fallot patients. Interact Cardiovasc Thorac Surg. 2011;12(6):987-92.
67. Haas F, Schreiber C, Hörer J, Kostolny M, Holper K, Lange R. Is there a role for mechanical valved conduits in the pulmonary position? Ann Thorac Surg. 2005;79(5):1662-7.
68. Shin HJ, Kim YH, Ko JK, Park IS, Seo DM. Outcomes of mechanical valves in the pulmonic position in patients with congenital heart disease over a 20-year period. Ann Thorac Surg. 2013;95(4):1367-71.
69. Miraldi F, Spagnesi L, Tallarico D, Di Matteo G, Brancaccio G. Sorin stentless pericardial valve versus Carpentier–Edwards Perimount pericardial bioprosthesis: Is it worthwhile to struggle? Int J Cardiol. 2007;118(2):253-5.
70. Lee C, Park CS, Lee CH, Kwak JG, Kim SJ, Shim WS, et al. Durability of bioprosthetic valves in the pulmonary position: long-term follow-up of 181 implants in patients with congenital heart disease. J Thorac Cardiovasc Surg. 2011;142(2):351-8.
71. Allen BS, El-Zein C, Cuneo B, Cava JP, Barth MJ, Ilbawi MN. Pericardial tissue valves and Gore-Tex conduits as an alternative for right ventricular outflow tract replacement in children. Ann Thorac Surg. 2002;74(3):771-7.
72. Morales DL, Braud BE, DiBardino DJ, Carberry KE, McKenzie ED, Heinle JS, et al. Perimount bovine pericardial valve to restore pulmonary valve competence late after right ventricular outflow tract repair. Congenit Heart Dis. 2007;2(2):115-20.
73. Fiore AC, Rodefeld M, Turrentine M, Vijay P, Reynolds T, Standeven J, et al. Pulmonary valve replacement: a comparison of three biological valves. Ann Thorac Surg. 2008;85(5):1712-8; discussion 8.
74. Kwak JG, Lee JR, Kim WH, Kim YJ. Mid-term results of the Hancock II valve and Carpentier-Edward Perimount valve in the pulmonary portion in congenital heart disease. Heart Lung Circ. 2010;19(4):243-6.
75. Shinkawa T, Anagnostopoulos PV, Johnson NC, Watanabe N, Sapru A, Azakie A. Performance of bovine pericardial valves in the pulmonary position. Ann Thorac Surg. 2010;90(4):1295-300.
76. Bowater S, Kerr R, Owen K, Hudsmith L, Clift P, Jones T, et al. Long term follow up perimount bioprosthetic pulmonary valves: a single centre's experience. Eur Heart J. 2011;32:328-9.
77. Chen XJ, Smith PB, Jaggers J, Lodge AJ. Bioprosthetic pulmonary valve replacement: Contemporary analysis of a large, single-center series of 170 cases. J Thorac Cardiovasc Surg. 2012:1461–6.
78. Shiokawa Y, Sonoda H, Tanoue Y, Nishida T, Nakashima A, Tominaga R. Pulmonary valve replacement long after repair of tetralogy of Fallot. Gen Thorac Cardiovasc Surg. 2012;60(6):341-4.
79. Ruzmetov M, Geiss DM, Fortuna RS. Outcomes of pericardial bovine xenografts for right ventricular outflow tract reconstruction in children and young adults. J Heart Valve Dis. 2013;22(2):209-14.
80. Aupart MR, Mirza A, Meurisse YA, Sirinelli AL, Neville PH, Marchand MA. Perimount pericardial bioprosthesis for aortic calcified stenosis: 18-year experience with 1133 patients. J Heart Valve Dis. 2006;15(6):768-75; discussion 75-6.
81. Marchand MA, Aupart MR, Norton R, Goldsmith IRA, Pelletier LC, Pellerin M, et al. Fifteen-year experience with the mitral Carpentier-Edwards PERIMOUNT pericardial bioprosthesis. Ann Thorac Surg. 2001;71(5):S236-S9.
82. Yacoub MH, Takkenberg JJ. Will heart valve tissue engineering change the world? Nat Clin Pract Cardiovasc Med. 2005;2(2):60-1.
83. Rippel RA, Ghanbari H, Seifalian AM. Tissue-engineered heart valve: future of cardiac surgery. World J Surg. 2012;36(7):1581-91.
84. Geva T. Tetralogy of Fallot repair: ready for a new paradigm. J Thorac Cardiovasc Surg. 2012;143(6):1305-6.
85. Kalfa D, Bacha E. New technologies for surgery of the congenital cardiac defect. Rambam Maimonides Med J. 2013;4(3):e0019.
86. Mack M. Progress toward tissue-engineered heart valves. J Am Coll Cardiol. 2014;63(13):1330-1.
87. Sutherland FWH, Perry TE, Sherwood M, Masuda Y, Garcia CA, McLellan DL, et al. Bone marrow derived pulmonary valve substitutes display favorable hemodynamic function in vivo. Circulation. 2002;106(19):658-.
88. Metzner A, Boldt J, Pohanke J, Fischer G, Schoettler J, Cremer J, et al. Comparison of Bone-Marrow Derived Cd133+-Cells and Cells Obtained from Carotid Artery after Percutaneous Tissue Engineered Pulmonary Valved Stent Implantation. J Am Coll Cardiol. 2012;59(13):E811-E.
89. Boldt J, Lutter G, Pohanke J, Fischer G, Schoettler J, Cremer J, et al. Percutaneous Tissue-Engineered Pulmonary Valved Stent Implantation: Comparison of Bone Marrow-Derived CD133+-Cells and Cells Obtained from Carotid Artery. Tissue Eng Part C-Me. 2013;19(5):363-74.
90. Walter EM, Sales VL, Sill B, Martin D, Rusk E, Emani S, et al. In vivo Implantation of a Functional Tissue Engineered Stentless Pulmonary Valve Using Bone-Marrow-Derived Mesenchymal Stem Cells and Circulating Endothelial Progenitor Cells. Circulation. 2010;122(21).
91. Hasan A, Ragaert K, Swieszkowski W, Selimovic S, Paul A, Camci-Unal G, et al. Biomechanical properties of native and tissue engineered heart valve constructs. J Biomech. 2014;47(9):1949-63.
92. Dohmen PM, Lembcke A, Holinski S, Pruss A, Konertz W. Ten years of clinical results with a tissue-engineered pulmonary valve. Ann Thorac Surg. 2011;92(4):1308-14.
93. Zaidi AH, Nathan M, Emani S, Baird C, Del Nido PJ, Gauvreau K, et al. Preliminary experience with porcine intestinal submucosa (CorMatrix) for valve reconstruction in congenital heart disease: Histologic evaluation of explanted valves. J Thorac Cardiovasc Surg. 2014.
94. Gilbert CL, Gnanapragasam J, Benhaggen R, Novick WM. Novel use of extracellular matrix graft for creation of pulmonary valved conduit. World J Pediatr Congenit Heart Surg. 2011;2(3):495-501.
95. Emmert MY, Weber B, Falk V, Hoerstrup SP. Transcatheter tissue engineered heart valves. Expert Rev Med Devices. 2014;11(1):15-21.
96. Neukamm C, Dohlen G, Lindberg HL, Seem E, Norgard G. Results 5 years after pulmonary valve replacement with a bovine pericardial valve. World Journal for Pediatric and Congenital Heart Surgery2014. p. 22.
97. Neukamm C, Try K, Norgård G, H. B. Right ventricular volumes assessed by echocardiographic three‐dimensional knowledge‐based reconstruction compared with magnetic resonance imaging in a clinical setting. Congenit Heart Dis. 2013;9(4):333-42.
98. Daubeney PEF, Blackstone EH, Weintraub RG, Slavik Z, Scanlon J, Webber SA. Relationship of the dimension of cardiac structures to body size: an echocardiographic study in normal infants and children. Cardiol Young. 1999;9(4):402-10.
99. Valente AM, Cook S, Festa P, Ko HH, Krishnamurthy R, Taylor AM, et al. Multimodality imaging guidelines for patients with repaired tetralogy of fallot: a report from the american society of echocardiography: developed in collaboration with the society for cardiovascular magnetic resonance and the society for pediatric radiology. J Am Soc Echocardiogr. 2014;27(2):111-41.
100. Freling HG, van Wijk K, Jaspers K, Pieper PG, Vermeulen KM, van Swieten JM, et al. Impact of right ventricular endocardial trabeculae on volumes and function assessed by CMR in patients with tetralogy of Fallot. Int J Cardiovasc Imaging. 2013;29(3):625-31.
101. Thompson WR, Gordon MD, Pescatello LS, editors. ACSM's guidelines for exercise testing and prescription. 8th ed. Baltimore, MD: Lippincott Williams & Wilkins; 2009.
102. Fredriksen PM, Ingjer F, Nystad W, Thaulow E. A comparison of VO2peak between patients with congenital heart disease and healthy subjects, all aged 8-17 years. Eur J Appl Physiol O. 1999;80(5):409-16.
103. Dagostino RB, Belanger A, Dagostino RB. A Suggestion for Using Powerful and Informative Tests of Normality. Am Stat. 1990;44(4):316-21.
104. Akins CW, Miller DC, Turina MI, Kouchoukos NT, Blackstone EH, Grunkemeier GL, et al. Guidelines for reporting mortality and morbidity after cardiac valve interventions. Eur J Cardiothorac Surg. 2008;33(4):523-8.
105. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet. 1986;1(8476):307-10.
106. Chen PC, Sager MS, Zurakowski D, Pigula FA, Baird CW, Mayer JE, Jr., et al. Younger age and valve oversizing are predictors of structural valve deterioration after pulmonary valve replacement in patients with tetralogy of Fallot. J Thorac Cardiovasc Surg. 2012;143(2):352-60.
107. Boethig D, Goerler H, Westhoff-Bleck M, Ono M, Daiber A, Haverich A, et al. Evaluation of 188 consecutive homografts implanted in pulmonary position after 20 years. Eur J Cardiothorac Surg. 2007;32(1):133-42.
108. Dearani JA, Danielson GK, Puga FJ, Schaff HV, Warnes CW, Driscoll DJ, et al. Late follow-up of 1095 patients undergoing operation for complex congenital heart disease utilizing pulmonary ventricle to pulmonary artery conduits. Ann Thorac Surg. 2003;75(2):399-411.
109. Baskett RJ, Nanton MA, Warren AE, Ross DB. Human leukocyte antigen-DR and ABO mismatch are associated with accelerated homograft valve failure in children: implications for therapeutic interventions. J Thorac Cardiovasc Surg. 2003;126(1):232-9.
110. Jashari R, Daenen W, Meyns B, Vanderkelen A. Is ABO group incompatibility really the reason of accelerated failure of cryopreserved allografts in very young patients?--Echography assessment of the European Homograft Bank (EHB) cryopreserved allografts used for reconstruction of the right ventricular outflow tract. Cell Tissue Bank. 2004;5(4):253-9.
111. Fuller S. Tetralogy of fallot and pulmonary valve replacement: timing and techniques in the asymptomatic patient. Semin Thorac Cardiovasc Surg Pediatr Card Surg Annu. 2014;17(1):30-7.
112. Gössl M, Holmes Jr DR. An update on transcatheter aortic valve replacement. Curr Prob Cardiology. 2013;38(7):245-83.
113. Kanter KR, Budde JM, Parks WJ, Tam VK, Sharma S, Williams WH, et al. One hundred pulmonary valve replacements in children after relief of right ventricular outflow tract obstruction. Ann Thorac Surg. 2002;73(6):1801-6; discussion 6-7.
114. Koh M, Yagihara T, Uemura H, Kagisaki K, Hagino I, Ishizaka T, et al. Long-term outcome of right ventricular outflow tract reconstruction using a handmade tri-leaflet conduit. Eur J Cardiothorac Surg. 2005;27(5):807-14.
115. Sekarski N, van Meir H, Rijlaarsdam ME, Schoof PH, Koolbergen DR, Hruda J, et al. Right ventricular outflow tract reconstruction with the bovine jugular vein graft: 5 years' experience with 133 patients. Ann Thorac Surg. 2007;84(2):599-605.
116. Sierra J, Christenson JT, Lahlaidi NH, Beghetti M, Kalangos A. Right ventricular outflow tract reconstruction: what conduit to use? Homograft or Contegra? Ann Thorac Surg. 2007;84(2):606-10; discussion 10-1.
117. Niemann PS, Pinho L, Balbach T, Galuschky C, Blankenhagen M, Silberbach M, et al. Anatomically oriented right ventricular volume measurements with dynamic three-dimensional echocardiography validated by 3-Tesla magnetic resonance imaging. J Am Coll Cardiol. 2007;50(17):1668-76.
118. Muller J, Engelhardt A, Fratz S, Eicken A, Ewert P, Hager A. Improved exercise performance and quality of life after percutaneous pulmonary valve implantation. Int J Cardiol. 2014;173(3):388-92.
119. Gaynor JW. Severe pulmonary insufficiency should be conservatively treated. Cardiol Young. 2005;15 Suppl 1:68-71.
120. Tateno S, Niwa K, Nakazawa M, Iwamoto M, Yokota M, Nagashima M, et al. Risk factors for arrhythmia and late death in patients with right ventricle to pulmonary artery conduit repair--Japanese multicenter study. Int J Cardiol. 2006;106(3):373-81.
121. Giannopoulos NM, Chatzis AC, Bobos DP, Kirvassilis GV, Tsoutsinos A, Sarris GE. Tetralogy of Fallot: influence of right ventricular outflow tract reconstruction on late outcome. Int J Cardiol. 2004;97 Suppl 1:87-90.
122. Oosterhof T, van Straten A, Vliegen HW, Meijboom FJ, van Dijk AP, Spijkerboer AM, et al. Preoperative thresholds for pulmonary valve replacement in patients with corrected tetralogy of Fallot using cardiovascular magnetic resonance. Circulation. 2007;116(5):545-51.
123. Therrien J, Provost Y, Merchant N, Williams W, Colman J, Webb G. Optimal timing for pulmonary valve replacement in adults after tetralogy of Fallot repair. Am J Cardiol. 2005;95(6):779-82.
124. Holmes KW. Timing of pulmonary valve replacement in tetralogy of fallot using cardiac magnetic resonance imaging: an evolving process. J Am Coll Cardiol. 2012;60(11):1015-7.
125. Norgard G, Gatzoulis MA, Moraes F, Lincoln C, Shore DF, Shinebourne EA, et al. Relationship between type of outflow tract repair and postoperative right ventricular diastolic physiology in tetralogy of Fallot. Implications for long-term outcome. Circulation. 1996;94(12):3276-80.
126. van der Hulst AE, Hylkema MG, Vliegen HW, Delgado V, Hazekamp MG, Rijlaarsdam ME, et al. Mild residual pulmonary stenosis in tetralogy of fallot reduces risk of pulmonary valve replacement. Ann Thorac Surg. 2012;94(6):2077-82.
127. Baumgartner H. What news in the 2010 European Society of Cardiology (ESC) guidelines for the management of grown-up congenital heart disease? J Cardiovasc Med (Hagerstown). 2013;14(2):100-3.
128. Ladouceur M, Gillaizeau F, Redheuil A, Iserin L, Bonnet D, Boudjemline Y, et al. Optimal follow-up in adult patients with congenital heart disease and chronic pulmonary regurgitation: towards tailored use of cardiac magnetic resonance imaging. Arch Cardiovasc Dis. 2013;106(1):27-35.
129. Moroseos T, Mitsumori L, Kerwin WS, Sahn DJ, Helbing WA, Kilner PJ, et al. Comparison of Simpson's method and three-dimensional reconstruction for measurement of right ventricular volume in patients with complete or corrected transposition of the great arteries. Am J Cardiol. 2010;105(11):1603-9.
130. Hoppe H, Derose T, Duchamp T, Mcdonald J, Stuetzle W. Surface Reconstruction from Unorganized Points. Comp Graph. 1992;26:71-8.
131. Quail MA, Frigiola A, Giardini A, Muthurangu V, Hughes M, Lurz P, et al. Impact of Pulmonary Valve Replacement in Tetralogy of Fallot With Pulmonary Regurgitation: A Comparison of Intervention and Nonintervention. Ann Thorac Surg. 2012;94(4): 1619–26.
132. Wald RM, Redington AN, Pereira A, Provost YL, Paul NS, Oechslin EN, et al. Refining the assessment of pulmonary regurgitation in adults after tetralogy of Fallot repair: should we be measuring regurgitant fraction or regurgitant volume? Eur Heart J. 2009;30(3):356-61.
133. Cui W, Kondo T, Anno H, Guo YY, Sato T, Sarai M, et al. The accuracy and optimal slice thickness of multislice helical computed tomography for right and left ventricular volume measurement. Chin Med J (Engl). 2004;117(9):1283-7.
134. Strugnell WE, Slaughter l R, Riley RA, Trotter AJ, Bartlett H. Modified RV short axis series--a new method for cardiac MRI measurement of right ventricular volumes. J Cardiovasc Magn Reson. 2005;7(5):769-74.
135. Kohler D, Arnold R, Loukanov T, Gorenflo M. Right ventricular failure and pathobiology in patients with congenital heart disease - implications for long-term follow-up. Front Pediatr. 2013;1:37.
136. van der Wall EE, Mulder BJ. Pulmonary valve replacement in patients with tetralogy of Fallot and pulmonary regurgitation: early surgery similar to optimal timing of surgery? Eur Heart J. 2005;26(24):2614-5.