|
|
||||||||
Ann Thorac Surg 2002;73:594-600
© 2002 The Society of Thoracic Surgeons
a University of Illinois at Chicago, The Heart Institute for Children, Hope Childrens Hospital, Oak Lawn, Illinois, USA
* Address reprint requests to Dr Ilbawi, The Heart Institute for Children, Hope Childrens Hospital, 4440 W 95th St, Oak Lawn, IL 60453, USA
Presented at the Thirty-seventh Annual Meeting of The Society of Thoracic Surgeons, New Orleans, LA, Jan 2931, 2001.
| Abstract |
|---|
|
|
|---|
Methods. The data on 12 patients who had the procedure between January 1989 and June 2000 were retrospectively reviewed. Associated lesions included ventricular septal defect in 12 patients, pulmonary stenosis in 10 patients, and moderate to severe tricuspid valve regurgitation in 4 patients. Mean age at operation was 9 ± 3.6 months. All patients had venous switch Mustard procedure. Tunneling of the morphologic left ventricle through the ventricular septal defect to the aorta with insertion of right ventricular to pulmonary artery conduit was performed in 10 patients, and arterial switch operation in 2. Concomitant tricuspid valvuloplasty was done in 2 patients and ventricular septal defect enlargement in 1.
Results. There was one hospital death (9%) in the patient who needed ventricular septal defect enlargement. Complications included atrioventricular block requiring pacemaker insertion in 1 patient (9%) and superior vena caval obstruction in 1 patient (9%). Follow-up is available on all patients 0.5 to 10 years (mean, 7.6 ± 3.1 years). All patients are asymptomatic. Exercise test results on the three oldest patients were normal. Bradytachyarrhythmias developed in 4 patients (36%). Right ventricular to pulmonary artery conduit replacement was needed in 5 patients 2.2 to 7.1 years (mean 5.2 ± 3.6 years) postoperatively. Mild to moderate tricuspid valve regurgitation persisted in 2 patients. Systemic left ventricular fractional shortening was 36% to 47% (mean, 39% ± 4.6%), and ejection fraction was 49% to 70% (mean, 60.8% ± 7.9%).
Conclusions. The double switch operation can be performed safely with minimal intermediate and long-term complications.
| Introduction |
|---|
|
|
|---|
Based on concerns related to poor performance of the right ventricle in the systemic position in this and other related lesions and the disappointing intermediate and long-term results of traditional repair of lesions associated with ventriculoarterial discordance, the concept of anatomic repair which incorporates the left ventricle in the systemic circulation was introduced with the hope of improving surgical results and longevity [13, 14].
Despite the relative complexity of this approach, its prompt adoption yielded excellent early results, low rates of operative mortality and morbidity, and encouraging early follow-up data, namely, good hemodynamic parameters and normal left ventricular function [1, 8, 11, 13, 15]. It has become an attractive alternative to the traditional surgical treatment because it aims at physiologic and anatomic repair of the lesion. Nevertheless, it has the drawbacks of utilizing a complication-prone venous switch operation and lacks the test of time and availability of long-term results [12].
In this report we review our experience with the procedure since inception and evaluate its intermediate results. Such evaluation is essential to determine whether the novel approach, with its inherent complexity, is better than the traditional way of managing congenitally corrected transposition and associated lesions.
| Material and methods |
|---|
|
|
|---|
|
All patients had preoperative 2-dimensional echocardiogram, Doppler studies, and cardiac catheterization. End-diastolic ventricular dimensions ranged between 107% and 169% of normal (mean, 143.0% ± 17.5%) for the right ventricle. Systemic oxygen saturation ranged from 78% to 94% (mean, 84.50% ± 4.7%). Left and right ventricular ejection fractions were 59% ± 3.9% and 54.9% ± 4.6% of predicted normal, respectively. The gradient across the pulmonary stenosis or band was between 41 and 77 mm Hg (mean, 64.8 ± 10.4 mm Hg), and distal and pulmonary artery pressure was between 15 and 33 mm Hg (mean 24.3 ± 7.9 mm Hg). Left ventricle to right ventricle pressure ratio was 1.0 in all patients except 1 who had a restrictive ventricular septal defect with the ratio being 0.8. The distance between the superior edge of the ventricular septal defect and aortic valve annulus varied between 0.7 and 2.4 cm. The aortic valve was always anterior with a variable degree of leftward orientation in relation to the pulmonary valve.
Technique
Repair was performed through a median sternotomy. A large piece of pericardium was harvested. Ascending aorta was cannulated cephalad after adequate mobilization. Bicaval cannulation was performed at the cavoatrial junction. Moderate to deep hypothermia (20°C to 24°C) was used. PH stat strategy was used during cooling and shifted to alpha stat strategy when the desired temperature was reached. Myocardial protection was achieved with antegrade blood cardioplegia for the venous switch operation and ventricular septal defect closure and with retrograde cardioplegia for the arterial switch operation.
The venous switch operation was performed using the Mustard procedure. The right atrium was opened obliquely and the incision extended through the atrial septum to the area between the two right pulmonary veins. The atrial septum was excised leaving a 3 to 5 mm margin superior and anterior to the coronary sinus. The roof of the coronary sinus was incised posteriorly into the left atrium and incorporated in the systemic venous channel. Autologous pericardium was used to construct the tunnel. Superficial and interrupted sutures were taken in the area anterior to the coronary sinus to avoid injury to the displaced atrioventricular node.
In patients with a normal left ventricular outflow tract the arterial switch operation was performed using the techniques generally utilized for D-transposition of the great arteries. A large button of aortic wall surrounding the coronary ostia was cut from the wall of the sinuses. The proximal coronary artery was mobilized to achieve tension-free coronary artery reimplantation. The coronary buttons were sutured to the pulmonary artery utilizing trap-door flaps. The LeCompte maneuver was possible in the 2 patients. The neopulmonary artery was reconstructed with autologous pericardium. The ventricular septal defect was closed through a right ventriculotomy (1 patient) or the neopulmonary valve (1 patient). Multiple interrupted sutures were placed on the right side of the septum to secure a 0.6-mm Gore-Tex patch (W. L. Gore & Associates, Flagstaff, AZ) in place and avoid the conduction system (Fig 1).
In patients with left ventricular outflow obstruction a right ventriculotomy was performed. A piece of 0.6-mm Gore-Tex patch was used to tunnel the left ventricular flow into the aorta using multiple interrupted sutures inferiorly and posteriorly and continuous sutures around the aortic annulus. In 1 patient the restrictive ventricular septal defect was enlarged inferiorly and caudally. The continuity between the right ventricle and pulmonary artery was established using a pulmonary homograft conduit 14 to 19 mm in diameter (mean, 16 ± 1.6 mm) placed to the right of the aorta in 8 patients and to its left in 2 (Fig 2). A tube extension of pericardium or descending aortic homograft was used to lengthen the pulmonary homograft when needed.
|
Data collection and analysis
All data were obtained from retrospective review of patients hospital charts and clinical records preoperatively and postoperatively. Left ventricular function and dimensions were obtained from serial echocardiographic evaluation. Volumes of the left ventricle were calculated from cineangiograms using the area-length method, and that of the right ventricle using Simpsons rule. Both were expressed as percentage of normal value for age. Exercise testing was performed using the standard Bruce exercise protocol for children. Data are presented as mean with standard deviation. Statistical analysis was performed utilizing paired or unpaired Students t test. A p value of 0.05 or less was considered significant.
| Results |
|---|
|
|
|---|
Intensive care unit stay of the survivors ranged between 3 and 11 days (mean, 5.3 ± 3.3 days) and hospital stay between 11 and 26 days (mean, 16.0 ± 5.0 days). All patients were on inotropic support and afterload reduction (dobutamine and milrinone) while in the intensive care unit. Two patients had prolonged chest tube drainage. Complete atrioventricular block requiring pacemaker insertion occurred in 1 patient (9%). All others were discharged in sinus rhythm.
Serial Holter and echocardiographic Doppler evaluations are available on all patients 0.6 to 10 years postoperatively (mean, 7.6 ± 3.1 years). All patients are asymptomatic (New York Heart Association functional class I-II). Bradytachyarrhythmias developed in 4 asymptomatic patients (36%), 2 with slow junctional rhythm and the other 2 with sick sinus syndrome, 3.7 to 7 years (mean, 5.1 ± 7 years) postoperatively. Medical treatment was needed in 2 and pacemaker insertion in 1 with resultant control of the arrhythmia. The remaining patients have sinus rhythm. Systemic left ventriclular fractional shortening was 36% to 47% (mean, 39% ± 4.6%) and ejection fraction 49% to 70% (mean, 60.8% ± 7.9%). Right ventricular ejection fraction ranged between 48% and 59% (mean, 53.8% ± 5.0%). There was no residual ventricular septal defect. Conduit replacement was performed in 5 patients 2.2 to 7.1 years postoperatively (mean, 5.2 ± 3.6 years) because of progressive stenosis with gradients between 29 and 76 mm Hg (mean, 42.6 ± 21.0 mm Hg) or moderate to severe regurgitation.
Tricuspid valve regurgitation improved to trivial in 2 patients and to mild to moderate in the other 2 patients. Exercise testing was performed on the oldest three patients (8.1 to 10.3 years old) 7 to 10 years postoperatively. All three demonstrated normal blood pressure elevation, heart rate response, and work capacity. No patient developed early or late subaortic stenosis.
| Comment |
|---|
|
|
|---|
Timing of the anatomic repair in patients with congenitally corrected transposition has evolved through the years and depends on symptomatology and the presence of associated lesions [15]. Asymptomatic patients with left ventricular outflow obstruction and ventricular septal defect maintain adequate left ventricular pressure and muscle mass and therefore can be electively operated on at 6 to 12 months of age, at which time the retrosternal space is adequate enough to accommodate a large right ventricular to pulmonary artery conduit without distortion of the conduit or compression of the coronary arteries. If the patient becomes significantly desaturated at a younger age, a palliative shunt is placed [2]. In patients with ventricular septal defect as the only associated lesion, neonatal anatomic repair is not recommended; instead, early banding is performed to protect the pulmonary vascular bed followed by the double switch operation at 6 months to 1 year of age. The delay of surgical intervention to this age promotes enlargement of the aortic sinuses, which in turn provides a larger button around the coronary ostia and consequent easier reimplantation of the coronary arteries at the time of the arterial switch operation.
Symptomatic patients with tricuspid regurgitation and right ventricular dysfunction are likely to be older. If the left ventricle is prepared in these patients, anatomic repair can be undertaken. Although the regurgitation tends to improve with time after the double switch operation, concomitant tricuspid valvuloplasty in selected cases can facilitate prompt recovery of right ventricular function. Symptomatic patients with unprepared left ventricle, conversely, may undergo pulmonary artery banding and simultaneous valve repair if regurgitation is present. This allows gradual recruitment of the left ventricle and interim hemodynamic stability.
Despite the strong data that support the application of the double switch operation to all patients with congenitally corrected transposition, there are a few anatomic variables that might decrease the attractiveness of the procedure or necessitate technical modification to optimize outcome. One such variable is hypoplasia of either ventricle [11, 15, 17]. A small left ventricle or mitral valve precludes the use of anatomic correction. A small right ventricle, although more common, is less critical. A pulsatile bidirectional Glenn Shunt or a small fenestration that allows limited right-to-left shunt at the atrial level could be used when the right ventricle is as small as 40% to 50% of predicted normal value. Another variable is ventricular septal defect size that is be too small to accommodate blood flow from the left ventricle to the aorta in patients undergoing the Rastelli procedure. Enlargement of a slightly restrictive defect caudally and inferiorly is feasible, but if extensive, it predisposes the patient to injury of the conduction system or to the septal perforators, leading to myocardial dysfunction and suboptimal results [1, 17, 19]. Therefore, patients with small defects who need a Rastelli procedure are best treated with the traditional approach. A third variable is coronary artery anatomy that may not lend itself to the arterial switch operation [20]. A lateral origin of the right coronary artery, distant from the pulmonary root, complicates the reimplantation of the artery into the neoaorta [15]. The use of a large aortic wall button and the pulmonary artery trap-door flap are technically helpful maneuvers. A pulmonary valve commissure may be too close to the area of coronary artery transfer [17]. The implantation of the coronary buttons cephalad to the deeply seated pulmonary valve solves this problem. Early bifurcation could make the transfer of coronary arteries more difficult. Limited proximal mobilization of the arteries decreases the chance of kinking or distortion. An additional anatomic variable is the extent of subaortic conus [16]. It is an important factor that determines the length and width of the tunnel necessary to direct left ventricular blood through the ventricular septal defect into the aorta. In most cases of congenitally corrected transposition the subaortic conus is attenuated, but occasionally it is long and hypertrophied, especially in older patients, requiring appropriate adjustment of tunnel location and size to prevent postoperative development of subaortic stenosis. In a few patients resection of portions of the large conus is needed, even at the expense of developing atrioventricular block [8]. Occasionally, construction of an adequate tunnel is complicated by chordal attachment of the tricuspid valve to the edge of the defect. Sacrificing some secondary chordae and tailoring the patch to accommodate primary chordal attachment provide ample tunnel width without compromising valve function.
The degree of leftward rotation of the aorta and its relationship to the pulmonary artery dictates the position of the right ventriculopulmonary artery conduit. An anterior and slightly leftward aorta is close to the midline and occupies most of the retrosternal space. Placement of a conduit to the right side of the aorta results in compression of the conduit and can accelerate dysfunction. It can also result in compression of the right coronary artery by the conduit and consequent myocardial ischemia. Therefore, it is ideal to place the conduit on the left side of the aorta whenever possible, although the homograft might need an extension to bridge the long distance between the pulmonary artery and right ventricle. Positioning of the conduit to the left of the aorta might not be necessary or feasible in patients with an aorta to the extreme left.
Tricuspid valve insufficiency in corrected transposition is a serious problem, and the anatomic features of the valve do not lend themselves readily to valvuloplasty techniques [3]. Repair of the valve in the past has been fraught with failure. Although anatomic correction decreases the severity of tricuspid valve regurgitation, occasionally patients may benefit from tricuspid valvuloplasty at the time of the switch operation [4, 18]. The use of tricuspid leaflet augmentation with autologous pericardium has been very successful in these patients and has resulted in prompt decrease in regurgitation and improved postoperative course.
The data strongly support the use of anatomic repair of congenitally corrected transposition of the great arteries in all patients who are symptomatic or have associated lesions, such as tricuspid valve regurgitation, ventricular septal defect, or pulmonary stenosis [1, 8, 13]. Extending the applicability of this procedure to asymptomatic patients with no associated lesions is justifiable but requires more objective data and further close scrutiny and assessment.
| Discussion |
|---|
|
|
|---|
My second comment is on the terminology. The title of your presentation says "double-switch operation," but in fact only a minority of your patients had a true double switch. Perhaps a more appropriate term might be "anatomic correction," because it does include two different operations to achieve anatomic repair. One is a true double switch and the other one is a Rastelli operation and an atrial switch.
DR ILBAWI: About the title, I do not know whether you can say that. The procedure involves double switching, because you are changing both the venous inflow and the arterial outflow. Whether you are doing it with an arterial switch directly or with a Rastelli procedure does not make a difference. It is just a matter of semantics if you like.
The issue of the venous switch operation in patients who have what we call cavoapical juxtaposition similar to what you are mentioning is a difficult one. I still feel that the Mustard procedure can be done in these patients. It takes a little bit more time and you have to pull a little bit more on the myocardium and the different structures, but I think it can still be done, and we have done it in 2 patients.
DR WILLIAM G. WILLIAMS: (Toronto, Ontario, Canada): Congratulations on a very nice series of difficult patients to manage. I think you have addressed a very important question about the place of this operation.
Our experience is very similar to yours, but the outcome has been not anywhere near as good, and we have had problems with atrioventricular block and with subaortic stenosis.
I think we should distinguish the double switch operation into two categories. Clearly the patients in whom you can do an arterial switch and Mustard procedure are ideal patients to manage with the so-called double switch. However, combining the complications of the Mustard procedure with those of the Rastelli procedure is certainly a lot less appealing, at least in my mind.
I noticed that your mean age was very much younger than ours. By operating early, you may have precluded some problems with the ventricular septal defects closing or excessive hypertrophy. So my first question is, can you comment on the elective age for doing a double-switch operation?
Second, it seems that a lot of these patients have small ventricular septal defects, if not restrictive, you worry about leaving them and having the child outgrow the ventricular septal defect at some later point as we have seen in several patients. Could you comment on enlarging the ventricular septal defect, and how you do that to both avoid atrioventricular block and avoid subsequent late obstruction?
DR ILBAWI: I think you are raising two very important points. When it comes to timing of the surgery, my feeling is that there is no place for neonatal intervention in this group of patients. I think this is a major operation and it is better done at a later age. We palliate all of these patients initially, and subsequently we do the double-switch or the Rastelli operation.
I do not like to wait too long because there is progressive hypertrophy of the myocardium, and that can interfere with myocardial protection as time goes on. So I choose anywhere between 6 and 14 months of age, depending on the onset of symptoms, to address the issue of double-switch or Rastelli and venous switch operation.
Delaying surgery a little bit is also important because it helps enlarge the aortic root and therefore helps you get a larger button around the coronary arties, which I consider very important for successful transplantation. It might also help in getting better exposure to the ventricular septal defect through the neopulmonary artery.
Now, the issue of ventricular septal defect is something that I am really becoming more and more concerned about as I learn more about this procedure. The ventricular septal defect in most of these patients is adequate early in life, but how do you judge it to be adequate for later on? Most of these patients have a left ventricular to right ventricular pressure ratio of 1. Their left ventricle has adequate muscle mass. So what are the criteria for deciding whether this ventricular septal defect is going to be adequate or not?
This is something for which I do not have the answer, but I am very concerned. If there is any question that the ventricular septal defect is restrictive, I prefer the traditional approach for the treatment of these patients rather than anatomic repair.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
M. G. Gaies, C. S. Goldberg, R. G. Ohye, E. J. Devaney, J. C. Hirsch, and E. L. Bove Early and Intermediate Outcome After Anatomic Repair of Congenitally Corrected Transposition of the Great Arteries Ann. Thorac. Surg., December 1, 2009; 88(6): 1952 - 1960. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Ly, E. Belli, B. Leobon, C. Kortas, O. E. Grollmuss, D. Piot, C. Planche, and A. Serraf Results of the double switch operation for congenitally corrected transposition of the great arteries Eur. J. Cardiothorac. Surg., May 1, 2009; 35(5): 879 - 884. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Sharma, S. Talwar, A. Marwah, S. Shah, S. Maheshwari, P. Suresh, R. Garg, B. S. Bali, R. Juneja, A. Saxena, et al. Anatomic repair for congenitally corrected transposition of the great arteries. J. Thorac. Cardiovasc. Surg., February 1, 2009; 137(2): 404 - 412.e4. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. W. Quinn, S. P. McGuirk, C. Metha, P. Nightingale, J. V. de Giovanni, R. Dhillon, P. Miller, O. Stumper, J. G. Wright, D. J. Barron, et al. The morphologic left ventricle that requires training by means of pulmonary artery banding before the double-switch procedure for congenitally corrected transposition of the great arteries is at risk of late dysfunction. J. Thorac. Cardiovasc. Surg., May 1, 2008; 135(5): 1137 - 1144.e2. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Shin'oka, H. Kurosawa, Y. Imai, M. Aoki, M. Ishiyama, T. Sakamoto, S. Miyamoto, K. Hobo, and Y. Ichihara Outcomes of definitive surgical repair for congenitally corrected transposition of the great arteries or double outlet right ventricle with discordant atrioventricular connections: Risk analyses in 189 patients J. Thorac. Cardiovasc. Surg., May 1, 2007; 133(5): 1318 - 1328. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. A. Alghamdi, B. W. McCrindle, and G. S. Van Arsdell Physiologic Versus Anatomic Repair of Congenitally Corrected Transposition of the Great Arteries: Meta-Analysis of Individual Patient Data Ann. Thorac. Surg., April 1, 2006; 81(4): 1529 - 1535. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Subtaweesin and S. Sriyoschati Early Results of Anatomic Repair in A Subgroup of Corrected Transposition Asian Cardiovasc Thorac Ann, September 1, 2005; 13(3): 208 - 210. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Dodge-Khatami, A. Kadner, F. Berger, H. Dave, M. I. Turina, and R. Pretre In the Footsteps of Senning: Lessons Learned From Atrial Repair of Transposition of the Great Arteries Ann. Thorac. Surg., April 1, 2005; 79(4): 1433 - 1444. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Aeba, T. Katogi, K. Koizumi, Y. Iino, M. Mori, and R. Yozu Apico-pulmonary artery conduit repair of congenitally corrected transposition of the great arteries with ventricular septal defect and pulmonary outflow tract obstruction: A 10-year follow-up Ann. Thorac. Surg., November 1, 2003; 76(5): 1383 - 1388. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. W. Duncan, R. B.B. Mee, C. I. Mesia, A. Qureshi, G. L. Rosenthal, S. G. Seshadri, G. K. Lane, and L. A. Latson Results of the double switch operation for congenitally corrected transposition of the great arteries Eur. J. Cardiothorac. Surg., July 1, 2003; 24(1): 11 - 20. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. J. Devaney, J. R. Charpie, R. G. Ohye, and E. L. Bove Combined arterial switch and Senning operation for congenitally corrected transposition of the great arteries: Patient selection and intermediate results J. Thorac. Cardiovasc. Surg., March 1, 2003; 125(3): 500 - 507. [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 |