| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2004;77:881-888
© 2004 The Society of Thoracic Surgeons
a Division of Cardiovascular Surgery, Rochester, MN, USA
b Division of Cardiovascular Diseases, Rochester, MN, USA
c Division of Pediatric Cardiology, Rochester, MN, USA
d Division of Biostatistics, Mayo Clinic and Mayo Foundation, Rochester, Minnesota, USA
* Address reprint requests to Dr Dearani, Mayo Clinic, 200 First St SW, Rochester, MN 55905, USA.
e-mail: jdearani{at}mayo.edu
Presented at the Thirty-ninth Annual Meeting of The Society of Thoracic Surgeons, San Diego, CA, Jan 31–Feb 2, 2003.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and methodsResultsCommentAcknowledgmentsDiscussionReferences |
|---|
METHODS: We retrospectively reviewed 102 patients (65 males, 37 females) who underwent operation with autologous tissue reconstruction ("peel operation") between May 1983 and November 2001, in which a prosthetic roof was placed over the fibrous bed of the explanted conduit. Ages ranged from 5 to 58 years old (median age 19 years old). Explanted conduits were Hancock (n = 54), homograft (n = 21), Tascon (n = 11), and other (n = 16). The conduit roof was constructed with pericardium (n = 91) and other (n = 11). A prosthetic pulmonary valve was utilized in 68 patients: porcine in 65 patients and mechanical in 3 patients. A nonvalved reconstruction was performed in 34 patients. Concomitant cardiac procedures were performed in 66 patients.
RESULTS: Early mortality overall was 2% (n = 2) and was 0% for patients who underwent isolated conduit replacement (n = 36). Mean follow-up was 7.6 years (maximum, 19 years). Overall survival at 10 and 15 years was 91% (84.7, 97.2) and 76% (62.8, 91.7), respectively. Nine patients required reoperation related to the peel operation: regurgitation in nonvalved conduit (n = 7); moderate pulmonary bioprosthesis stenosis and regurgitation with atrial arrhythmia (n = 1); and pulmonary bioprosthesis endocarditis (n = 1). Overall survivorship free of reoperation for peel reconstruction failure at 10 and 15 years was 90.7% (82.6, 99.6) and 82% (69.4, 97.0), respectively. Survivorship free of reoperation for patients with a prosthetic valve was 93.7%, and for those with no prosthetic valve was 80.0% at 15 years (p = 0.57). At late follow-up, 89% of patients were in New York Heart Association functional class I or II.
CONCLUSIONS: The peel operation simplifies conduit replacement, can be performed with low risk, and provides a generous-sized flow pathway. In our experience late results demonstrate a lower freedom from reoperation than conventional prosthetic or homograft conduits.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentAcknowledgmentsDiscussionReferences |
|---|
It is widely accepted that the ideal conduit has not yet been developed. Late problems include conduit degeneration with progressive obstruction or regurgitation, lack of growth, potential for infection, and the need for anticoagulation when a mechanical valve is used. The need for one or more reoperations for conduit replacement remains inevitable for most patients.
These late problems prompted us to explore a different strategy when conduit replacement was required. A new technique was developed which requires an autologous tissue reconstruction utilizing a pericardial roof over the fibrous bed of the explanted conduit [8–10], with or without insertion of a prosthetic pulmonary valve. Early results of this procedure, now known as the "peel operation," have been reported previously [9].
The purpose of this study was to examine the long-term outcome of all patients at our institution with congenital heart defects who underwent replacement of a PV-PA conduit with a peel operation with or without the insertion of a prosthetic pulmonary valve. We evaluated patient survival, functional status, durability of the peel reconstruction, and need for subsequent reoperation.
|
Patients and methods |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentAcknowledgmentsDiscussionReferences |
|---|
|
At the time of the index peel operation, it was the second sternotomy for 46 patients (45%), the third sternotomy for 44 patients (43%), and a fourth, fifth or sixth sternotomy for 12 patients (12%). The sizes of the explanted conduits ranged from 8 to 26 mm (mean 22 mm, median 22 mm) and were Hancock (n = 54), homograft (n = 21), Tascon (n = 11; Medtronic, Inc), Dacron graft (n = 6), Ionescu-Shiley (n = 4; Shiley Inc, Irvine, CA), Carpentier-Edwards (n = 4; Edwards Lifesciences, Irvine, CA), and Gore-Tex graft (n = 2; W.L. Gore & Associates, Flagstaff, AZ). In this series, 92 of the explanted conduits were valved and 10 were nonvalved. The intervals between prior conduit operation and peel operation ranged from 1.0 to 28.1 years (mean 11.2 years, median 9.9 years). The total number of conduits that had been implanted in each patient before the peel operation ranged from 1 to 5 (mean 1.4).
Indications for conduit replacement in patients with a valved conduit included symptomatic patients with a peak systolic gradient of greater than 40 to 50 mm Hg, or a senescent bioprosthesis in a patient who required reoperation for an unrelated cardiac diagnosis. Indications for conduit replacement in asymptomatic patients were pulmonary ventricular pressure approaching systemic pressure, progressive pulmonary ventricular dilatation or dysfunction, the progression of pulmonary atrioventricular (predominantly tricuspid) valve regurgitation, and deterioration in exercise test performance. For patients with nonvalved conduits without obstruction, the indications for conduit replacement for symptomatic and asymptomatic patients were the same, with the exception of the PV pressure and the PV-PA gradient criteria.
Preoperative conduit gradients and degree of regurgitation were evaluated by echocardiography or cardiac catheterization. Gradients across the peel reconstruction were obtained intraoperatively by direct measurement or transesophageal echocardiography. Transthoracic echocardiography was performed routinely in patients before hospital discharge.
The surgical technique of the peel operation has been described and illustrated in previous reports [9, 10]. Repeat sternotomy can be challenging because the conduit and valve ring or calcified homograft may be adherent to or eroding into the sternum. Although central cannulation is preferred, femoral artery and vein cannulation may be necessary to establish cardiopulmonary bypass before or during resternotomy. When femoral cannulation is required, sufficient dissection is performed after resternotomy to allow conversion to central cannulation. Pressure measurements in the cardiac chambers, conduit, and distal pulmonary arteries are obtained if feasible. Aortic and right atrial cannulation with mild to moderate hypothermia without aortic occlusion is employed when there are no atrial or ventricular septal defects, and the heart is allowed to be perfused and beating throughout the operation. Bicaval cannulation and aortic occlusion are utilized for concomitant intracardiac procedures that require cardioplegic arrest such as repair of atrial or ventricular septal defects and aortic valve replacement. Topical iced saline is also used whenever feasible to optimize myocardial protection.
After establishment of cardiopulmonary bypass and placement of an aortic tack vent, a vertical incision is made over the anterior aspect of the obstructed conduit. The conduit is dissected free from the external fibrous peel and excised (Fig 1). The fibrous sides and floor of the conduit bed are maintained and preserved. Calcified homografts are managed by leaving most of the posterior wall of the homograft intact and by endarterectomizing the lateral edges so that the adventitia of the homograft and adjacent fibrous tissue are available for suture placement. The right and left pulmonary arteries are measured with probes in order to identify any stenoses. The pulmonary ventricular outflow tract is examined and any muscular or fibrous obstruction is resected as indicated. Concomitant intracardiac procedures are then performed.
|
Glutaraldehyde-preserved bovine pericardium is currently utilized for construction of the roof of the new conduit. The distal suture line and the shape of the patch are altered as necessary to provide relief of any right or left pulmonary artery stenoses. The suture line is restricted to the fibrous peel at the edge of the conduit bed to avoid injury to adjacent, obscured coronary arteries. Insertion of a bioprosthetic valve is now routine. In this series, we implanted a large-sized stented porcine bioprosthesis (usually 29 or 31 mm in the adult). The valve is positioned so as to minimize compression by the sternum; in patients with transposition of the great arteries (TGA), truncus arteriosus, and corrected TGA, placement is usually close to the distal anastomosis (ie, pulmonary artery confluence). In other patients, especially those with pulmonary atresia or tetralogy of Fallot, the valve can often be placed at the level of the pulmonary annulus or just proximal to it in the pulmonary ventricular outflow tract. The valve is usually secured with mattress sutures that are buttressed with felt pledgets posteriorly and laterally in the conduit bed. Anteriorly, the sewing ring is attached to the pericardial roof with continuous monofilament suture. Care is taken to ensure the pericardium is configured so there is free passage through the reconstruction distal to the valve struts, in order to prevent obstruction at that level. The reconstruction is completed by sewing the proximal pericardial patch to the pulmonary ventricular outflow tract (Fig 1).
After patient rewarming and discontinuation of cardiopulmonary bypass, pressures are routinely measured in the pulmonary ventricle and distal pulmonary artery. Assessment by intraoperative transesophageal echocardiography is also performed. Partial sternectomy or costal cartilage resection are performed as needed to prevent compression of the reconstruction with sternal closure.
Failure of the conduit constructed by the peel technique was defined as need for reoperation for PV-PA pathway obstruction, regurgitation (especially in nonvalved conduits), and pulmonary valve endocarditis.
Follow-up status of the patients was determined principally by review of patient records and by written questionnaires to the patient or families and, when needed, to the local physicians. Telephone interviews were conducted with patients who did not respond to the written questionnaire. Data requested included functional status, the occurrence of cardiovascular events, and any reoperation or hospitalization. If the patient had died, the death certificate or medical records of the hospital and physician were reviewed. This study was approved by the Institutional Review Board, and patient or parent informed consent was obtained.
Early mortality was defined as death occurring within 30 days of operation or at any time during the index hospitalization. The probability of survival was estimated by the Kaplan-Meier method, and 95% confidence intervals were calculated for each estimate. Survival curves of the patients were compared with the expected curves of persons of the same age and gender, as derived from vital statistics for the West North Central region of the United States. The statistical significance of observed versus expected survival and freedom from reoperation were assessed with a one-sample log-rank test. Statistical significance of survival curves comparing categorical variables used a two-sample log-rank test. Values of p less than 0.05 were considered statistically significant.
|
Results |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentAcknowledgmentsDiscussionReferences |
|---|
Femoral arterial cannulation was performed in 35 patients (34%); concomitant femoral venous cannulation was performed in 15 patients (15%). Cardiopulmonary bypass time ranged from 34 to 438 min (mean 143 min, median 130 min). Aortic occlusion was required in 41 patients; cross-clamp times ranged from 10 to 273 min (mean 77 min, median 68 min). Concomitant cardiac procedures were performed in 66 patients (Table 2).
|
Overall early mortality for the peel operation was 2% (n = 2); it was 0% for the 36 patients undergoing isolated conduit replacement. One early death occurred early in the series (1985) in a patient whose conduit had eroded into a single left coronary artery that was unroofed during conduit removal. The coronary artery was repaired, a nonvalved peel reconstruction was made, residual atrial and ventricular defects were closed, and a right pulmonary artery stenosis was relieved, but the patient subsequently died of biventricular failure. The second early mortality occurred in a patient with a single pulmonary artery and pulmonary vascular obstructive disease who died from progressive pulmonary hypertension and right ventricular failure after a valved peel reconstruction with concomitant patch enlargement of a superior vena caval stenosis, closure of an atrial septal defect, and cryoablation of right atrial isthmus for atrial flutter.
Nonfatal morbidity occurred in 35 patients (34%) and included: transient arrhythmias in 15 patients (atrial in 8 patients, ventricular in 7 patients); reoperation for bleeding in 5 patients; heart block requiring permanent pacemaker in 2 patients; and stroke, renal failure requiring dialysis, sternal infection, and lower extremity compartment syndrome secondary to an intraaortic balloon pump in 1 patient each. One patient who had a preoperative history of ventricular tachycardia received an implantable cardioverter defibrillator before hospital dismissal.
The use of a prosthetic valve with the peel reconstruction did not increase early mortality (1.5% with vs 2.9% without valve, p = 0.9) or early morbidity (33.8% with vs 35.2% without valve, p = 0.8). The hospital stay for the early survivors ranged from 4 to 40 days (mean 9 days, median 7 days).
There were 11 late deaths. The mean time to late death was 7.4 years (maximum 15.3 years). Causes of late mortality were cardiac in 7 patients, noncardiac in 1 patient (passenger in motor vehicle accident), and unknown in 3 patients. Cardiac causes of death were sudden in 3 patients, heart failure associated with pulmonary hypertension in 3 patients, and adult respiratory distress syndrome 7-days after reoperation in 1 patient. Mean follow-up of all 89 late survivors was 7.6 years (maximum 19 years). Overall survival for all 102 patients at 5, 10, and 15 years was 91% (84.7, 97.2), 91% (84.7, 97.2), and 76% (62.8, 91.7), respectively (Fig 2). There was no significant difference in late survival between patients who received a valve versus those who did not (p = 0.8).
|
The mean time to reoperation after the peel operation was 9.5 years. Survival free from reoperation for failure of peel reconstruction at 5, 10, and 15 years was 97.6% (94.2, 100), 90.7% (82.6, 99.6), and 82% (69.4, 97.0), respectively (Fig 3). Survival free from reoperation for peel reconstruction failure in patients with a prosthetic pulmonary valve at 5, 10, and 15 years was 98.4% (95.3, 100), 93.7% (84.7, 100), and 93.7% (84.7, 100), respectively. Survival free from reoperation for peel reconstruction failure in patients without a prosthetic pulmonary valve at 5, 10, and 15 years was 97% (91.3, 100), 90.0% (79.8, 100), and 80.0% (65.3, 98.0), respectively. The difference in Kaplan-Meier survival free from reoperation of 93.7% at 15 years when a prosthetic valve was used in the repair (2 reoperations in 67 early survivors) versus 80.0% at 15 years when no prosthetic valve was used (7 reoperations in 33 early survivors). The value p = 0.57 was not statistically significant because of the small sample sizes at 15 years. Survival free from reoperation for peel reconstruction failure stratified according to diagnostic categories is illustrated in Figure 4.
|
|
|
Comment |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentAcknowledgmentsDiscussionReferences |
|---|
Our attempt to reduce the incidence of late conduit failure was the construction of an autologous tissue conduit, with or without a valve. Advantages of this technique include an autologous tissue floor with a pericardial roof that does not form obstructive tissue peels, and the diameter of the pathway can be as large as desired, allowing a large bioprosthesis to be inserted. A large bioprosthesis decreases the PV-PA gradient, decreases turbulence at the valve, and may increase longevity of the valve. A low gradient across the reconstruction becomes of greater importance when the pulmonary ventricle is dilated and has significant dysfunction. Additionally, the distal portion of the pericardial patch can be tailored to address and proximal right or left pulmonary artery stenoses.
We have examined the freedom from reoperation for conduit failure in an age-matched group of patients who have received a Hancock conduit, a homograft conduit, or a valved peel reconstruction (Fig 5). The peel operation had statistically significant better freedom from reoperation compared with the homograft (p = 0.001). Although the peel operation had more favorable durability than the Hancock conduit, this did not reach statistical significance (p = 0.19) due to the small numbers in the peel operation group at late follow-up.
|
These data do not demonstrate with statistical significance that survival free of reoperation for failure of nonvalved peel reconstructions is superior to that for valved reconstructions, due in part to the small number of reoperations in both groups. Perhaps our current preference for valved reconstructions has been unduly influenced bt the 7 patients with nonvalved reconstructions who required reoperation for developing right heart failure. Longer late survival data in both groups are necessary to clarify this issue.
Three patients (2.9%) received a mechanical pulmonary valve. Two of these patients had pulmonary hypertension. One patient died from pulmonary hypertension and congestive heart failure 15-years after operation. The two remaining patients are alive and well. Although the use of mechanical valves in the pulmonary position has been described [26, 27], we have generally avoided their use because of low right-sided pressures, potential thrombosis problems, and the need for long-term anticoagulation. In addition, we have now been able to demonstrate good success in eliminating atrial arrhythmias, often present and necessitating anticoagulation in many of these patients, by performing a concomitant right-sided maze procedure [28]. Because of the favorable durability of stented porcine bioprostheses noted in this review and others [24], we currently prefer this prosthesis when a pulmonary valve insertion is performed.
The peel operation is based on the fibrous tissue that forms around an implanted extracardiac conduit. Dacron conduits incite the greatest fibrous reaction. Woven Dacron (eg, Hancock conduit) is least attached to the surrounding peel, and the conduit can often be freed by blunt dissection. Knitted Dacron with collagen impregnation (eg, Tascon conduit) is usually tightly incorporated in the fibrous tissue and must be freed by sharp dissection; at times it may be expedient to leave the posterior wall of the conduit in situ to avoid injury to underlying coronary arteries or other important structures. Calcified homografts can be excised anteriorly and endarterectomized laterally to provide attachment for the pericardial roof. Occasionally the location of the conduit to be explanted does not allow a peel reconstruction. For example, this might pertain to a patient with corrected transposition, pulmonary atresia, and isolated dextrocardia in whom an extended conduit arises from the pulmonary (morphologically left) ventricular apex and courses through the right side of the mediastinum/pleural space to insert superiorly into the pulmonary artery confluence.
Our current approach to the selection of primary PV-PA conduits is to use a pulmonary homograft during the neonatal and infant periods up to 3 to 4 years old, in the absence of pulmonary hypertension. When pulmonary hypertension is present, or a small pulmonary homograft is unavailable, an aortic homograft is selected. Beyond 4 years of age, we prefer a porcine-valved conduit. When reoperation is required for conduit failure, we would proceed with peel operation with insertion of a porcine bioprosthesis. The polytetrafluoroethylene monocusp technique [29] and early results have been reported [30]. We have utilized this technique in infants and children when a transannular reconstruction of the right ventricular outflow tract is required at initial operation. We have also had encouraging results in the early postoperative period, however, the subsequent development of pulmonary regurgitation occurs in the majority of patients. Because right ventricular dilatation and dysfunction is usually present to some degree at the time of reoperation, we prefer to place a bioprosthetic valve during the reconstruction in order to maintain a competent valve for as long as possible.
The search continues for the ideal extracardiac conduit. Although the need for reoperation is inevitable for most patients, the risk of reoperation is low and the majority of patients enjoy a good quality of life. At present, the peel operation provides the most favorable freedom from reoperation for conduit failure and is our operation of choice when conduit replacement is required.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentAcknowledgmentsDiscussionReferences |
|---|
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentAcknowledgmentsDiscussionReferences |
|---|
Today, the Mayo Clinic group has summarized their results in 102 patients treated by the peel operation approach since May 1983. The results are impressive. The operative mortality is low and the patients seem to do well for at least a decade following operation. I find that their evolving approach to this operation fits many of my own biases. Their technique is quite similar to our approach. I also note that they are routinely inserting bioprosthetic valves in all conduit replacement patients utilizing this technique. This approach may not be appropriate for all patients.
Because first generation porcine valves usually failed rapidly in children and young adults, I have always felt that a porcine valve should not be inserted in a child unless it is absolutely necessary. In this series, most of the patients are young adults. The median age at the time of peel operation was 19 years old. It would be helpful to know how many patients under the age of 10 underwent this operation. The actuarial curves look quite impressive, however, the mean follow-up of all 89 late survivors is only 7.6 years. There are relatively few patients in the valve group at 10 years and 15 years. Since the overall series is front-end loaded with valveless reconstructions, it is possible that degeneration of the bioprosthetic valves within the next 4 or 5 years will change the long-term analysis. I will be looking forward to the next analysis in about 5 years.
I also believe that more than 10 years follow-up will be necessary for most patients before we can make substantial inferences about the long-term performance of any bioprosthesis in a child or young adult. We are all optimistic that the latest generation of porcine valves, stented or nonstented, as well as the pericardial valve will provide longer service in young patients. However, it is my impression that there is very little hard data supporting the improved long-term survival of these valves in children and young adults.
In summary, patients with a requirement for a competent pulmonary valve should have a bioprosthetic valve implanted. The Mayo Clinic technique seems ideally suited for that subset of patients. On the other hand, it is my guess that all patients undergoing implantation of a bioprosthetic valve will eventually require re-replacement. In the past, I had hoped that homograft valves would be the answer. Unfortunately, homograft valves in the right heart have not lived up to their early promise. A small number of my patients do have homograft valves that are continuing to function well beyond 10 years, suggesting that in some patients a homograft can be a long-term solution. Maybe the new generation of decellularized homograft valves or bioengineered valves will provide better long-term performance. In any event, I still try to avoid implanting xenograft valves in patients under the age of 15 years old unless they have a requirement for a guaranteed competent valve for the next 5 to 10 years. Many patients still do well without a pulmonary valve if right ventricular function is normal and the pulmonary vascular bed is unobstructed.
I have several questions for the authors. How many children less than 10 or 12 years of age have undergone this approach and what has happened to the bioprostheses implanted in these younger patients? Do you have any echocardiographic long-term follow-up data in the group of patients that underwent implantation of a bioprosthetic valve? Specifically, are these patients beginning to manifest any evidence of early degeneration of the valves 7 to 10 years following implantation? Can we anticipate that the present generation of bioprosthetic valves will function well for 20 years in this population? If not, are we not committing most of these relatively young patients to another operation?
I would like to thank the Society for allowing me the privilege of discussing this important contribution to the congenital heart surgery literature.
DR BERMUDEZ: Thank you, Dr Lamberti, for your comments. Of the 102 patients, 8% of patients were 10 years of age or younger and 30% of the patients were 15 years of age or less.
There were 2 reoperations for failure of a valved peel reconstruction. The first was in a 14-year-old who developed endocarditis 1 year after peel operation. The second was in a 17-year-old who underwent an arrhythmia operation and conduit replacement 6 years after peel operation.
Echocardiographic data were available at late follow-up (mean 5.4 ± 4.3 years). The mean peak systolic gradient across the right ventricular outflow tract was 25 mmHg and the mean right ventricular systolic pressure was 55 mmHg. There was no significant difference in these variables when valved and nonvalved reconstructions were compared. Severe conduit regurgitation was obviously present in the nonvalved reconstructions. Three patients with a valved reconstruction had severe regurgitation, 1 had moderate regurgitation, and the remainder had mild to none.
In general, our results with the use of the homograft in the right ventricular outflow tract have been disappointing. Cryopreserved aortic homografts typically fail from stenosis. However, pulmonary homografts fail more rapidly from pulmonary regurgitation, especially when pulmonary artery pressures are elevated. Despite the overall poor durability of homografts in this position, we continue to favor their use in infants and small children because of the favorable handling characteristics. We have previously reported good durability of the Hancock porcine-valved Dacron conduit in older children and young adults. In addition, the Dacron conduit provides the ideal fibrous bed after explantation for subseqent peel operation.
It is not conclusive from these data whether a pulmonary valve prosthesis is necessary or preferable in a child at the time of conduit replacement since these data do not demonstrate with statistical significance that survival free of reoperation for failure of valved peel recontructions is superior to that for nonvalved reconstructions. This is due in part to the small number of reoperations in both groups. Perhaps our current preference for valved reconstructions has been unduly influenced by the 7 patients with nonvalved reconstructions who required reoperation for developing right heart failure. Longer late survival data in both groups are necessary to clarify this issue.
|
References |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentAcknowledgmentsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. Mohammadi, E. Belli, I. Martinovic, L. Houyel, A. Capderou, J. Petit, C. Planche, and A. Serraf Surgery for right ventricle to pulmonary artery conduit obstruction: risk factors for further reoperation Eur. J. Cardiothorac. Surg., August 1, 2005; 28(2): 217 - 222. [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 |
