| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2006;81:671-677
© 2006 The Society of Thoracic Surgeons
a Department of Cardiovascular Surgery, National Cardiovascular Center, Osaka, Japan
b Department of Cardiothoracic Surgery, Royal Brompton Hospital, London, United Kingdom
Accepted for publication August 15, 2005.
* Address correspondence to Dr Yagihara, Department of Cardiovascular Surgery, National Cardiovascular Center, 5-7-1 Fujishiro-dai, Suita, Osaka, 565-8565, Japan (Email: yagihara{at}hsp.ncvc.go.jp).
Presented at the Forty-first Annual Meeting of The Society of Thoracic Surgeons, Tampa, FL, Jan 24–26, 2005.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsCommentDISCUSSIONReferences |
|---|
METHODS: A total of 45 patients underwent the double-switch operation. Their ages ranged from 6 months to 21 years. Associated malformations included pulmonary atresia in 27, pulmonary stenosis in 11, and Ebstein's malformation in 5. An atrial switch plus an arterial switch procedure was performed in 7, and an atrial switch plus a Rastelli-type ventriculoarterial switch procedure in 38. Follow-up ranged from 6 months to 15 years.
RESULTS: Early mortality was 8.9% (n = 4). In the latter half of the series (n = 23, since 1994), there was no early death. Six patients died late. Actuarial survival at 5 and 10 years was 83.6% and 77.6%, respectively. Six patients required conduit replacement, and 2 required revision of an intraatrial baffle for pulmonary venous channel obstruction and infection, respectively. Freedom from reoperation was 95.3% at 5 years and 76.2% at 10 years. Freedom from arrhythmia was 88.8% at 5 years and 78.4% at 10 years. The systemic ventricular ejection fraction was 0.568 ± 0.103 at 1 year (n = 39), 0.555 ± 0.105 at 5 years (n = 17), and 0.539 ± 0.098 at 10 years (n = 12).
CONCLUSIONS: The surgical results of the double-switch operations have been improving. Intermediate follow-up suggests that these alternative procedures are a reasonable option for patients with atrioventricular discordance.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDISCUSSIONReferences |
|---|
Since 1987, anatomic biventricular repair using the double-switch operations (DSO) has been our principal choice for patients with AVD [5]. The DSO includes two combinations of procedures. A combined atrial and arterial switch procedure can be performed in those patients in whom the pulmonary valve is normal and left ventricular pressure has been maintained at systemic levels by virtue of a large ventricular septal defect or previous pulmonary artery banding. In those patients with pulmonary valve stenosis or atresia associated with a large ventricular septal defect, a combined atrial switch and a Rastelli-type ventriculoarterial switch procedure can be performed. These alternative procedures result in support of the systemic circulation by the morphologic left ventricle and the mitral valve. The tricuspid valve, even with some morphologic abnormalities, is placed within the low-pressure pulmonary circulation and is likely to function reasonably. This approach should improve the long-term outcome of patients with AVD [6–10]. However, the DSO might result in future complications, such as atrial arrhythmia or venous pathway obstruction related to the surgical maneuver of intraatrial redirection of blood, or obstruction of the conduit. It has not yet been determined whether the potential advantages outweigh the risks. In this study, we evaluated the intermediate results of the DSO to clarify the optimal surgical strategy in this setting.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDISCUSSIONReferences |
|---|
|
In 12 patients with normal pulmonary valves, the left ventricular pressure was maintained at a systemic level in 3 owing to a large amount of shunt associated with the ventricular septal defect. In the other 9 patients, left ventricular pressure was depressed and tricuspid regurgitation was present. We performed pulmonary artery banding in 4; 3 were banded for congestive heart failure, and 1 for left ventricular training, followed by a combined atrial and arterial switch procedure. The remaining 5 were older patients who developed significant tricuspid regurgitation and congestive heart failure (4, 14, 21, 37, and 52 years). We carried out tricuspid valve replacement in these 5 patients [11].
A total of 45 patients underwent the DSO; a combined atrial and arterial switch procedure in 7, and a combined atrial switch and a Rastelli-type ventriculoarterial switch procedure in 38. The age at time of operation ranged from 6 months to 21 years (Fig 2).
|
Previous Surgery
Four patients underwent pulmonary artery banding as stated above. One of them had coarctation of the aorta, and extended direct anastomosis of the aorta concomitant with pulmonary artery banding was required in his early infancy. A total of 36 systemic to pulmonary artery shunts were created in 21 patients. Six with hypoplasia of the pulmonary artery, probably related to the previous shunting operation, required pulmonary artery reconstruction using a heterologous pericardial roll by means of a lateral thoracotomy [12]. Two patients had a major aortopulmonary collateral artery and arborization anomaly. They underwent staged unifocalization of the pulmonary artery [13].
Surgical Procedure
We previously described the details of our surgical procedure [5]. Standard aortobicaval cardiopulmonary bypass with moderate hypothermia (28°C) and cold antegrade crystalloid cardioplegia were used.
Atrial Switch Procedure
The Mustard procedure was performed in 36 patients and the Senning procedure in 9. In the Mustard procedure, a glutaraldehyde-treated heterologous pericardial patch was used for an intraatrial baffle in 32 and an expanded polytetrafluoroethylene patch in 4. To supplement the pulmonary venous chamber in the Mustard procedure, as is occasionally required in the Senning procedure, we sutured an additional patch to the old right atrium.
Arterial Switch Procedure
Seven patients with a normal pulmonary valve underwent an arterial switch procedure. A standard Lecompte maneuver was applied in 4. The original Jatene procedure was applied in the remaining 3, who had a relatively side-by-side relationship of the great arteries. For reconstruction of the pulmonary artery, an autologous or heterologous pericardial patch was generously inserted in all.
Rastelli-Type Ventriculoarterial Switch Procedure
Thirty-eight patients with pulmonary valve stenosis or atresia underwent a Rastelli-type procedure for ventriculoarterial switching. The internal conduit was patched to create a channel between the morphologic left ventricle and the aortic orifice. The intraventricular communication was enlarged to create an unobstructed left ventricular outflow tract in 5. In 4 patients, who had relatively small ventricular septal defects, an additional aortopulmonary anastomosis was constructed [5, 14, 15].
The right ventricular outflow tract was reconstructed with an external conduit in 26; a handmade trileaflet conduit made of glutaraldehyde-treated heterologous pericardium (what we call a "valved pericardial roll") was used in all [16]. Nonconduit repair, minimizing the use of prosthetic material, was carried out in 12. A direct anastomosis between the right ventricle and the pulmonary artery was created in 4; an REV technique was applied to facilitate direct anastomosis in 2 [17, 18]. A piece of autologous pericardial patch was inserted posteriorly between the right ventriculotomy and the pulmonary artery in 8, creating continuity of autologous tissue. For the anterior aspect of the right ventricular outflow tract, an expanded polytetrafluoroethylene patch was used in 5, a heterologous pericardium in 3, and an autologous pericardium in 4. In 11 of 12 patients, we avoided the use of a valved conduit; a monocusp made of autologous pericardium or expanded polytetrafluoroethylene sheet was applied.
Data Analysis
All surviving patients had periodic follow-up at our institution. Patients' data were retrospectively reviewed. The follow-up period ranged from 6 months to 15 years, with a mean of 6.4 ± 4.8 years. Routine postoperative catheterization studies were carried out in the majority at approximately 1, 5, and 10 years after the operation.
Data are presented as the mean ± the standard deviation. Cumulative survival estimates were made by the Kaplan-Meier method. The log-rank test was applied for comparison of survival curves. Differences in continuous variables were assessed by Student's t test. Values of p less than 0.05 were considered significant. All statistical tests were conducted with JMP 5.1.1 software (SAS Institute Inc, Cary, NC).
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDISCUSSIONReferences |
|---|
|
|
No patient developed obvious left ventricular outflow tract obstruction after a Rastelli-type ventriculoarterial switch procedure. Among those 7 patients who underwent the arterial switch procedure, none have developed significant aortic regurgitation, despite 4 who required pulmonary artery banding before the DSO. A total of 8 patients have needed reoperation thus far (Fig 5).
|
|
|
|
Comment |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDISCUSSIONReferences |
|---|
Selection of candidates is essential to accomplish the DSO successfully. The DSO combined with an atrial and arterial switch procedure is attractive in patients with a normal pulmonary valve. However, particularly in older patients, the left ventricle placed in the low-pressure circulation for a prolonged period is not prepared to support the systemic circulation. Pulmonary artery banding appears to be an option for providing adequate left ventricular training when done at an early age; however, it is not always indicated for older patients. In this series, only 1 young patient (5 years old) with depressed left ventricular pressure underwent pulmonary artery banding with the aim of left ventricular training followed by a combined atrial and arterial switch procedure 4 months later. The other 3 patients required pulmonary artery banding in their early infancy mainly because of congestive heart failure associated with tricuspid regurgitation. Considering the difficulties with left ventricular training, we abandoned the idea of the DSO for the 4 older patients, although they had a reasonable morphology for this alternative procedure.
The optimal timing of the DSO is also crucial. For patients with a normal pulmonary valve, we prefer surgery in late infancy to minimize the deterioration of the left ventricular musculature and the deleterious effect of banding on the native pulmonary valve that is to be placed in the aortic position. We have no experience with this procedure in the neonatal period or early infancy. When we encounter a symptomatic neonate or baby in early infancy, we consider pulmonary artery banding primarily, followed by the DSO several months later.
We believe that a good indication for the DSO combined with an atrial and a Rastelli-type ventriculoarterial switch procedure was pulmonary stenosis or atresia with a ventricular septal defect that was subaortic in position and suitable for rerouting with an intraventricular baffle from the left ventricle to the aorta. Occasionally, particularly in patients with mild pulmonary stenosis, the ventricular septal defect was unsuitable for rerouting with respect to size or location. This situation might be treated by direct enlargement of the ventricular septal defect or creation of an additional aortopulmonary anastomosis [5, 14, 15]. With these maneuvers, no patient has developed significant left ventricular outflow tract obstruction after ventriculoarterial switching. We previously reported that excessive enlargement of the intraventricular communication might induce ventricular dysfunction, particularly when the ventricular septal defect is a perimembranous inlet type [5, 20]. The presence of such a restrictive ventricular septal defect might be better treated by construction of an additional aortopulmonary anastomosis. An aortic translocation or a truncal switching might be a possible option to establish ventriculoarterial switching in such a situation [21, 22].
Regarding the timing of surgery for patients with pulmonary valve stenosis or atresia, we prefer to perform the DSO somewhat later. Operation in infancy in this setting might be disadvantageous because the ventricular volume is often too small at this stage of life to construct an unobstructed intraventricular channel. However, in older patients, there is another problem of palliative systemic to pulmonary shunting that might cause a volume load on the heart. Moreover, an acquired systemic to pulmonary collateral artery might develop in patients with prolonged cyanosis, which is unequivocally a risk of the operation [23]. It might cause both undesirable volume loading of the ventricle and unexpected airway or intestinal bleeding during the postoperative period. In the early era of this series, 3 patients experienced sudden bleeding in the airway during the postoperative period. These 3 older patients, 8, 9, and 21 years of age, underwent multiple systemic to pulmonary shunting operations as palliative procedures.
Sophistication of the surgical strategy should contribute to improvement of early results of the DSO. There is no doubt that better results were also related to technical improvement [5, 20]. The perfusion time was shorter in the latter half (317 ± 95 versus 259 ± 47 minutes; p = 0.014), as well as the cross-clamping time (166 ± 32 versus 147 ± 34 minutes; p = 0.058).
Reinterventions in the intermediate term are common, as prosthetic material is often required to accomplish the DSO. In particular, valved conduits inevitably deteriorate and obstruct within the intermediate term. We initially preferred the use of an external conduit for right ventricular outflow tract reconstruction in a Rastelli-type procedure. In this setting, we always placed the conduit to avoid crossing the midline to prevent compression of the conduit between the heart and the sternum [5]. The conduit tended to be long and winding, and early obstruction was a concern. Moreover, in the reoperation, dissection of the conduit placed far from the midline might be complicated. However, in our experience, conduit life was reasonable in comparison with other kinds of conduit repair, and there was no particular problem with conduit replacement [16]. Since 1992, nonconduit repair using autologous tissue has been our primary technique for infants and young children. This technique might reduce the requirement for reoperation, although right ventricular dysfunction owing to pulmonary regurgitation might adversely influence the outcome [24]. It is encouraging that all patients in this series having nonconduit repair have been free from reintervention. Currently, we reserve the use of conduits for older children and adults, particularly for patients with elevated pulmonary vascular resistance or ventricular dysfunction who might be helped by competent valve function [16].
In the early era, we primarily chose the Mustard procedure for atrial switching. This technique is easily applicable to various situations, including apicocaval juxtaposition or anomalous venoatrial connection associated with heterotaxy [5, 25]. However, calcification of the heterologous pericardium, placed in the atrium in the Mustard procedure, occasionally caused venous pathway obstruction. In our series, 10 patients required balloon dilatation of the narrowed venous pathway. We generally decided on early catheter intervention, even when the narrowing was still mild and subclinical, considering that this technique was often ineffective once the baffle had severely calcified. Actually, only 3 of these 10 patients had symptoms of venous pathway obstruction. Currently, we prefer the Senning procedure, in which whole autologous tissue is used for the intraatrial rerouting. We expect this technique might minimize long-term venous pathway obstruction.
Aortic regurgitation after an arterial switch procedure is a major concern over the long term, because pulmonary artery banding is common in this setting. In our series, the aortic valves functioned well in all, including 4 patients who required pulmonary artery banding before the DSO.
Atrial tachyarrhythmia is the most frequent morbid problem. It is an inherent complication of the atrial switching procedure, although this particular anomaly is often accompanied by atrial rhythm disturbances before surgery [1–3]. An effort to avoid atrial arrhythmia is mandatory, including meticulous design of the baffle and careful attention to the suturing technique.
Catheterization data showed that function of the left ventricle placed in the systemic circulation has been reasonably maintained in the majority. We previously reported that the coronary veins of the morphologic left ventricle commonly drain into the pectinated floor of the right atrium directly, not to the coronary sinus [26]. Such independent openings of the small coronary veins might be injured at the time of atrial switching, resulting in impairment of ventricular function during the early postoperative period; however, we could not find any such deleterious effects in the intermediate term.
In conclusion, the surgical results of the DSO have been improving. The majority of survivors is doing well with respect to New York Heart Association functional class, and is free from medications. We believe that the DSO is able to minimize deleterious complications associated with AVD.
|
DISCUSSION |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDISCUSSIONReferences |
|---|
I am particularly interested in the patients with dextrocardia, to give us a little bit of an insight into the two technical aspects. One is the Mustard/Senning part of the operation and the other one the placement of the conduit.
In our small experience with patients with dextrocardia is that the performance of the Mustard operation is quite difficult due to the small size of the right atrium. In these patients we have performed a modified Senning operation by using the in situ pericardial well technique.
My last question is, based on your experience, are you ready to offer a double-switch operation in a neonate with corrected transposition and intact ventricular septum, since left ventricular preparedness is not an issue in the first few weeks of life.
DR KOH: Your first question is regarding the atrial switching in the setting of dextrocardia.
In the earlier era, we preferred the Mustard procedure in the patient with apicocaval juxtaposition. Those patients have a rather small atrial wall, as you mentioned, and we think the Mustard procedure is easily applicable. The Senning procedure is rather difficult in this setting.
Recently we preferred the Senning procedure in order to minimize future venous pathway obstruction. In the Senning procedure, generally we augment the functional left atrium with a Gore-Tex patch.
And what is your next question?
DR TCHERVENKOV: Is the placement of the conduit in patients with dextrocardia to the right or the left of the aorta?
DR KOH: We always place the conduit not to cross the midline, on the opposite side of the caval veins. We intend to avoid compression between the heart mass and the sternum as well as turbulence at the anastomosis. This design generally makes the conduit long and winding; however, our data showed conduit life was satisfactory.
DR TCHERVENKOV: And my last question, what would you recommend for the 3-week-old neonate with corrected transposition and intact ventricular septum? Would you recommend the double-switch procedure?
DR CARL L. BACKER (Chicago, Illinois): He doesn't have enough time to answer that one, Christo.
DR TCHERVENKOV: All I need is a yes or a no.
DR KOH: No. Initially we will place banding, and 6 or 5 months later we will carry out the double-switch operation.
DR GLEN S. VAN ARSDELL (Toronto, Ontario, Canada): This is a fascinating paper with a beautiful, detailed analysis, but I think it doesn't completely address a very important question that we, as a society, still have to deal with.
First off, I think there's a lot of interest in the VSD (ventricular septal defect). How do you objectively evaluate your VSD to insure that it's of adequate size?
I believe your group has earlier shown that if you enlarge the VSD the ventricular function is worse. So how do you deal with the VSD? And are you following that? What type of gradients do you have? Have you had to reoperate for VSDs?
DR KOH: In the patient with pulmonary atresia or pulmonary stenosis, VSD is always located at the subaortic position and is large enough. When pulmonary stenosis is mild, VSD may be small. In such cases, we carried out VSD enlargement or Damus-Kaye-Stanzel anastomosis, what we call an additional aortopulmonary anastomosis, to create an unobstructed ventricular channel. We think excessive enlargement of VSD may impair ventricular function due to damage of the septal branches. One patient in our early series developed ventricular dysfunction after VSD enlargement, as you mentioned; however, no patient among the other 4 who required VSD enlargement in this study had similar results. For the patients with small VSD, additional aortopulmonary anastomosis is an effective alternative to create an unobstructed channel. There is no case that developed LVOTO (left ventricular outflow tract obstruction) in our series.
DR VAN ARSDELL: And maybe I should just ask one rhetorical question. Yesterday we talked about Rastellis, and the outcome for Rastellis, in the best-case scenario, is era-independent and 60% survival at 20 years. I would submit, as a rhetorical question, our Fontan survivals at 20 years are probably about 85% or 90%. Should we, as a society, be exploring doing Fontans on these patients instead of pursing a biventricular repair?
DR BACKER: That sounds like a good topic for next year's debates, and we'll put you as one of the protagonists.
DR VAN ARSDELL: I don't have the answer.
DR BACKER: Well, that's why we're going to have you talk.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDISCUSSIONReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  B. Sareyyupoglu, H. M. Burkhart, D. J. Hagler, J. A. Dearani, A. Cabalka, F. Cetta, and H. V. Schaff Hybrid approach to repair of pulmonary venous baffle obstruction after atrial switch operation. Ann. Thorac. Surg., November 1, 2009; 88(5): 1710 - 1711. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. P. Graham Jr The Year in Congenital Heart Disease J. Am. Coll. Cardiol., July 24, 2007; 50(4): 368 - 377. [Full Text] [PDF]
|
|
|
|
|
|
T. P. Graham Jr The Year in Congenital Heart Disease J. Am. Coll. Cardiol., June 20, 2006; 47(12): 2545 - 2553. [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 |
