| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Division of Cardiothoracic Surgery, St. Louis University School of Medicine, Cardinal Glennon Children's Hospital, St. Louis, Missouri
b Indiana University School of Medicine, James Whitcomb Riley Children's Hospital, Indianapolis, Indiana
Accepted for publication November 21, 2007.
* Address correspondence to Dr Fiore, St. Louis University School of Medicine, Cardinal Glennon Children's Hospital, 1465 S Grand Blvd, St. Louis, MO 63104 (Email: fiorem2{at}slu.edu).
Presented at the Fifty-third Annual Meeting of the Southern Thoracic Surgical Association, Tucson, AZ, Nov 8–11, 2006.
| Dr Brown discloses that he has a financial relationship with Medtronic.
|
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
Methods: From January 1995 to August 2006, 82 patients (mean age, 22.7 years) underwent valve replacement with porcine (49 patients), bovine pericardial (18 patients), or pulmonary homograft (15 patients) prosthesis at a mean of 15.3 years after initial outflow tract reconstruction. Excluded were patients with extracardiac conduits, monocusp valves, or the Ross procedure. The groups were similar with respect to age, body surface area, degree of regurgitation, right ventricular dimension, right ventricular to pulmonary artery gradient, and valve size. Follow-up was longer in the homograft cohort (porcine, 20 ± 27 months; pericardial, 42 ± 21; homograft, 49 ± 40; p < 0.01).
Results: All three prostheses significantly reduce chronic pulmonary regurgitation, but late insufficiency was higher with homografts. Right ventricular dimension was significantly reduced in the stented but not the allograft cohorts. Late valve dysfunction was highest with homografts (54%), followed by porcine (19%) and pericardial valves (5.5%; p < 0.05. Functional class and mild to moderate tricuspid insufficiency significantly improved with pulmonary valve replacement. Early and late mortality was 3.6% and 1.2%, respectively.
Conclusions: All three prostheses performed similarly for 3 years. Pulmonary regurgitation developed more frequently in homografts albeit at a longer duration of follow-up.
Chronic pulmonary insufficiency is a common problem after intervention to relieve pulmonary stenosis whether it is accomplished surgically as in transannular repair of tetralogy of Fallot or employing catheter-based intervention. The consequences of longstanding pulmonary regurgitation (PR) include right ventricular dilatation, tricuspid regurgitation, diastolic and systolic biventricular dysfunction, prolongation of the QRS duration, arrhythmias, and sudden cardiac death [1]. Although the majority of patients with severe PR remain asymptomatic, 5% to 30% requires pulmonary valve replacement (PVR) to relieve the deleterious effects of chronic pulmonary insufficiency on ventricular function and exercise capacity [2]. Numerous reports have demonstrated the importance of a competent pulmonary valve to relieve symptoms and preserve cardiac performance [3–5]. However, controversy remains with respect to the optimal time to intervene with valve replacement and the best prosthesis to implant. At the present time, options include mechanical as well as several biological valves. The stented xenograft valve, Medtronic mosaic porcine (PO) (Medtronic Inc, Minneapolis, MN), the Carpentier-Edwards bovine pericardial valve (PE) (Edwards Lifesciences, Irvine, CA), and the pulmonary homograft (HO) (CryoLife Inc, Kennesaw, GA) are three such commonly employed bioprostheses for PVR and form the basis of this report. Our purpose was to retrospectively compare the clinical and hemodynamic performance of these three biological valves in patients requiring pulmonary valve replacement for chronic pulmonary insufficiency.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
|
The indications for pulmonary valve replacement included one or more the following: (1) asymptomatic patients with severe PR and evidence of progressive right ventricle (RV) dilatation or dysfunction, or both; (2) symptomatic patients with longstanding severe PR and RV dilatation with or without RV dysfunction; (3) asymptomatic or symptomatic patients with moderate to severe PR in association with moderate to severe tricuspid insufficiency; (4) moderate to severe PR associated with RV or left ventricle dysfunction, or both; (5) moderate to severe PR and RV dilatation associated with serious ventricular arrhythmias especially if the QRS duration exceeded 180 ms. An additional diagnostic criteria for PVR is RV end-systolic volume two times or more than the left ventricular end-systolic volume as determined by cardiac magnetic resonance imaging [6].
Valve failure was defined as the need for explantation, and valve dysfunction was defined as a peak echocardiographic gradient across the prosthesis of 40 mm Hg or greater or severe PR, or both. All valves requiring explantation were dysfunctional.
Early mortality is defined as death in the hospital or within 30 days of discharge. All other mortality is considered late.
Patients, families, cardiologists, and primary care physicians were contacted when necessary to acquire current follow-up information. Follow-up was achieved in 78 hospital survivors. The pulmonary homograft valve was the first prosthesis inserted in this patient population and was associated with a significantly longer mean follow-up: PO, 20 ± 27 months; PE, 42 ± 21 months; HO, 49 ± 40 months (p < 0.01). The interval from PVR to latest echocardiographic follow-up was recorded for all patients: PO, 25.0 ± 27 months; PE, 41 ± 26 months; HO, 46 ± 36 months (p < 0.01). All patients were placed on aspirin therapy postoperatively.
Operative Technique
The type of valve used for PVR was chosen at the surgeon's discretion without randomization. The PVR and tricuspid valve repairs were performed with the heart beating using bicaval cannulation and mild hypothermia (32°C to 34°C). If aortic cross clamping was employed to close residual atrial and ventricular septal defects, myocardial protection was by cold intermittent blood cardioplegia.
The cardiopulmonary bypass time was not significantly different among the three cohorts: PO, 136 ± 49 minutes; PE, 132 ± 44; HO, 157 ± 17 (p = not significant). Twenty-five patients required aortic cross clamping to close atrial (18 patients; 22%), or ventricular septal defects (4; 5%), or both (2; 2.4%). Mitral valve replacement was required in 1 patient. Branch pulmonary artery reconstruction (16 patients; 19%) was performed with the heart beating. No significant differences in the performance of concomitant procedures were noted among the three cohorts.
The transannular patch or infundibular chamber was opened and the residual patch material with the native valve tissue resected. The posterior one third of the stented valve annulus was sewn to the native pulmonary valve annulus with continuous polypropylene suture (Prolene; Ethicon, Sommerville, New Jersey). Anteriorly, a diamond-shaped patch of Gore-Tex (W.L. Gore & Assoc, Flagstaff, Arizona) or bovine pericardium was used to roof the right ventricular outflow tract over the prosthesis. Patients receiving a pulmonary homograft had the outflow end of the prosthesis sewn end to end to the divided main pulmonary artery, and the inflow portion sewn to the infundibular incision, supplemented anteriorly with a patch of bovine pericardium or Gore-Tex.
Severe tricuspid regurgitation was repaired using either a DeVega annuloplasty or by obliterating the posterior leaflet using a continuous horizontal mattress stitch with pledgets along the posterior leaflet annulus [7, 8]. If necessary, additional annuloplasty sutures at the anterior septal and posterior septal commissures were employed to create a competent bicuspid valve.
Gore-Tex membrane pericardial substitute was used in all patients. The size of the prosthetic pulmonary valve inserted was not significantly different among the three cohorts: PO, 25 ± 2.7 mm; PE, 24 ± 2.6 mm; HO, 23 ± 4.8, mm (p = not significant), but the indexed effective orifice area was larger for the homograft cohort (PO, 1.2 cm2/M2; PE, 1.3 cm2/M2; HO, 3.9 cm2/M2; p < 0.05).
Statistics
Categorical data are presented as percentages accompanied by the number of cases. Continuous data are presented as medians with ranges and means with standard deviations as appropriate. Comparisons between groups were performed by one-way analysis of variance, Fisher's exact test, or 
2 as appropriate. Continuous characteristics and outcomes were compared using t tests. Time from initial valve insertion to repeat valve reimplantation was analyzed using Kaplan-Meier analysis. A p value of less than 0.05 was considered significant. All analyses were performed using the statistical package for the social sciences (SPSS) software (SPSS, Chicago, Illinois).
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
|
|
|
|
|
Valve Explantation
Valve explantation was required in 8 of 78 long-term survivors (4 HO; 3 PO; 1 PE; 10%). Four homografts required explantation at 17 and 22 months for acute insufficiency and at 72 and 78 months for chronic insufficiency and stenosis (HO, 4 of 13; 31%). One PO valve (21 mm) was explanted at 36 months secondary to staphylococcal endocarditis, and two additional PO valves became stenotic and explanted at 33 (21 mm) and 42 months (19 mm), (3 of 47; 6%). These three valves were implanted in children 14, 6, and 5 years of age, respectively. The final explantation occurred 30 months postoperatively in a patient in whom endocarditis developed after a dental procedure (PE, 1 of 18; 5%).
The actuarial freedom from explantation is shown in Figure 5. At 5 years, the actuarial freedom from explantation was 92% for the PE cohort, but fell to 78% for the PO and HO groups. At 6 years, the freedom from explantation of the HO valve was 35%. The population sizes are too small to allow meaningful statistical comparison.
|
Mortality
There were 3 early deaths (3 of 82; 3.6%). A 14-year-old boy with Down's syndrome, tetralogy of Fallot, and complete atrioventricular canal with severe biventricular dysfunction underwent mitral and PVR (PO) with tricuspid valve repair; death occurred secondary to ventricular dysfunction and multiorgan system failure 7 days postoperatively while on mechanical support. A second death occurred in a 4-year-old girl with pulmonary valve stenosis and insufficiency after tetralogy of Fallot repair; she had ventricular ectopy at the time of PVR (HO) and experienced sudden unexplained death at home 1 month postoperatively. The remaining early death was of a 38-year-old woman with tetralogy of Fallot, residual ventricular septal defect, and severe RV dysfunction. She underwent homograft PVR, septal defect closure, and tricuspid valve repair. Initial recovery was uneventful, but on postoperative day 5, sudden hypotension and ventricular fibrillation developed. Postmortem examination demonstrated RV dilatation without evidence of infarction or pulmonary emboli. One late death in a 10-year-old boy with Down's syndrome was secondary to bilateral atypical pneumonia 6 weeks after porcine valve replacement (1 of 79; 1.2%).
Heart transplantation was required in 1 patient with severe biventricular dysfunction 23 months after PVR (PO).
Tricuspid Insufficiency
Tricuspid valve repair was performed concomitantly in 19 patients (19 of 82, 23%) in whom severe insufficiency at the time of PVR was identified. Among the 78 long-term survivors, 60 patients (60 of 78, 77%) had tricuspid incompetence less than severe and underwent only PVR. The degree of tricuspid regurgitation in all patients was reassessed at a mean follow-up of 33 months. In patients receiving only PVR, the level of tricuspid regurgitation had significantly decreased presumably secondary to the reduction in right ventricular dimension after PVR alone (tricuspid regurgitation before PVR 1.3 ± 0.6 versus after PVR, 0.4 ± 0.5; p < 0.01). Among 17 survivors having concomitant PVR and tricuspid valve repair, the degree of regurgitation was reduced at latest follow-up but the difference was not significant (before PVR 2.9 ± 0.7 versus after PVR 1.9 ± 0.4; p = not significant).
|
Comment |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
An area of controversy surrounds the optimal prosthesis to implant. In this report, we compared the clinical and hemodynamic performance of the pulmonary homograft to that of two stented valves, the Medtronic mosaic and the Carpentier/Edwards bovine pericardial prostheses. Homografts have been preferred by some since they are available in small sizes requiring no anticoagulation and have low gradients. Pulmonary homografts are preferable because they have less elastic tissue and a lower amount of calcium than aortic homografts [9]. In this series, the lowest peak echocardiographic right ventricle to pulmonary artery gradient was seen in the homograft cohort. However, homografts deteriorate over time because of calcification and valve insufficiency. Poor performance of allografts for right ventricular outflow tract obstruction has recently been confirmed by Brown and coworkers [10] in 117 patients for whom the 5-year actuarial freedom from failure was 60%.
In our report, the pulmonary homograft demonstrated comparatively higher degrees of dysfunction secondary to early valve insufficiency and freedom from reoperation at 6 years of 35%. The early allograft insufficiency may have contributed to the increased right ventricular dimension we observed in this cohort. The HO valve is not inserted orthotopically because the inflow end of the conduit is placed extra anatomic and contains a roofing patch. Turbulent flow or leaflet distortion in this location may in part account for early failure. The extended durability of the homograft valve in the true orthotopic position has been confirmed in patients having the Ross procedure for whom the 5-year freedom from homograft explantation is 80% to 90% [11].
The Medtronic mosaic porcine valve is a newer generation stented valve. It is glutaraldehyde fixed, undergoes zero pressure fixation, and is treated with the antimineralization agent amino oleic acid. This agent binds covalently to the bioprosthetic tissue through amino linkages on residual aldehyde groups and inhibits the influx of calcium from circulating blood onto the leaflet cusps. Amino oleic acid is effective in attenuating cusp, but not aortic wall calcification in animals [12]. This feature seemed attractive and enhanced our enthusiasm for using this valve in children and young adults. However, at a mean follow-up of only 20 months, we observed the highest peak RV to pulmonary artery gradient across this prosthesis. The porcine valve exhibited significantly less regurgitation and dysfunction when compared with homografts, but the follow-up was significantly shorter for the porcine cohort. Kanter and associates [13] reported a favorable experience with the porcine valve. They compared stented porcine and homografts in 100 patients having PVR and demonstrated an actuarial freedom from reoperation of 100% at 8 years for the porcine and 70% for the homograft cohort. They further demonstrated that in children less than 3 years of age, most of whom received homografts, the actuarial freedom from reoperation at 8 years was 39% compared with 100% for patients beyond 3 years of age in whom the porcine prosthesis was more commonly employed [13].
The bovine pericardial valve in which the mean follow-up was 42 months appeared to be the most advantageous. This cohort experienced the lowest comparative rate of valve dysfunction and a trend toward lower actuarial freedom from valve explantation. In this report, the actuarial freedom from valve explantation (Fig 5) must be interpreted cautiously as the number of patients at risk is small and no statistical comparison could be performed. Allen and coworkers [14] reported favorable results in 48 patients receiving the pericardial valve followed for a mean of 43 months. Freedom from reoperation was 100% at 5 years [14]. Our follow-up is comparatively short for all three prostheses. Important clinical differences will not likely become apparent until 8 to 10 years of follow-up.
The occurrence of tricuspid insufficiency with chronic PR and RV dilatation is a marker for RV function. We believe, as others do, that it may be hazardous to delay PVR until after the onset of moderate to severe tricuspid regurgitation [15]. This excessive right ventricular chronic volume overloading and subsequent ventricular dysfunction that can occur when both right sided valves are severely insufficient is a strong impetus to proceed with earlier PVR. In this report, 2 of 3 hospital deaths secondary to severe biventricular dysfunction occurred in 19 patients in whom tricuspid valve repair for severe tricuspid regurgitation was required (19 of 82; 23%). Moreover, we observed that among 60 patients with less than severe tricuspid regurgitation, if PVR is performed early, then the reduction in RV dimension will be accompanied by a regression in tricuspid regurgitation [16].
In patients requiring PVR, the effective orifice area of the selected prosthesis is critical. The minimally acceptable effective orifice area for an aortic valve replacement is greater than 0.85 cm2/m2 [17]. For PVR, we believe it should be 1 to 1.3 cm2/m2 or greater. Biologic prostheses mounted in a frame or a stent will always have an effective orifice area one third to one half lower than the equivalent size stentless valve. In this report, the stented porcine had a lower indexed effective orifice area than the stented pericardial valve, and a corresponding greater degree of valve dysfunction. Therefore, if a stented prosthesis is desired, but the target effective orifice area cannot be achieved in that patient, then a stentless valve such as the porcine freestyle or the bovine jugular vein (Contegra) remain highly attractive alternatives to homografts [18, 19].
Limitations
This study is limited by several important factors. The three cohorts were not prospectively randomized, and valve selection was at each individual surgeon's preference without regard for anatomy. Although the utilization of each prosthesis was distributed equally between both institutions, there was a greater numerical bias toward the porcine valve, especially in the last 5 years. The overall follow-up was relatively short, and the length of follow-up for each group was unequal, with a longer interval among the homograft and pericardial cohorts.
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
Do you have the ages of the patients who had their valves replaced? Did they tend to be the younger patients and not the older ones? That goes with my first comment about the older patients; my thought was that they would be more protected from valve deterioration.
DR FIORE: The porcine valves explanted were in younger children, 8 to 10 years of age. They were 19- and 21-mm valves. The effective orifice area of a 19- or 21-mm prosthesis is probably not large enough for a child that size. I believe there was patient-valve mismatch. Histologically, the leaflet architecture of the explanted prostheses looked very good, and so we tend to believe that it was a patient-valve mismatch problem.
DR KANTER: Earlier this week at the postgraduate course, one of your coauthors, Dr Brown, extolled the virtues of the Contegra valve. As this was a consecutive series, were there no Contegra valves used in this series, or were you reserving them just for right ventricular to pulmonary artery conduits?
DR FIORE: The bovine jugular vein is, we think, a promising conduit for right ventricular outflow tract reconstruction, and can be used as a pulmonary valve replacement. Insertion of a Contegra would of course require division of the main pulmonary artery. It is also important to remember that unfortunately the device is under HDE category. One of the criteria for using this prosthesis under those auspices is that it cannot be inserted in someone over 18 years of age. Therefore, it is not an alternative for older patients.
I should comment, however, that the largest Contegra is 22 mm. If you think about it, a 22-mm Contegra, which is stentless, has an effective orifice area of 3.8 cm2. Therefore, the effective orifice area of the largest Contegra would be more than adequate for a growing teenager. However, it can only be inserted in patients under 18 years of age.
DR KANTER: You didn't tell us in your presentation or in your paper at what time postoperatively the follow-up echocardiograms were performed. Certainly, valve dysfunction is time related. Can you give us an idea as to when each of these valves started to develop pulmonary insufficiency or pulmonary stenosis?
DR FIORE: That is a good question, Kirk. If I understand your question correctly, you would like me to define the interval of time when we first noticed dysfunctionality in each of the three cohorts as judged echocardiographically. The answer to that question is certainly in the database. I will have to review that and plan to include the information in the manuscript.
DR KANTER: That can then give us an idea as to when they started deteriorating. Finally, can you give us an idea of the cost of the three valves? Would that enter into the equation?
DR FIORE: With respect to cost, the Contegra is slightly cheaper than a homograft. Currently, a pulmonary homograft is approximately $10,000, whereas a Contegra is approximately $4,000. The cost of a porcine freestyle valve is $8,700, whereas the cost of a Medtronic mosaic porcine or a bovine pericardial valve is $6,700.
DR KANTER: Again, congratulations on an excellent presentation.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  V. Dayan, F. Gutierrez, L. Cura, G. Soca, and A. Lorenzo Two Cases of Pulmonary Homograft Replacement for Isolated Pulmonary Valve Endocarditis. Ann. Thorac. Surg., June 1, 2009; 87(6): 1954 - 1956. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
