Eur J Cardiothorac Surg 2002;21:773-779
© 2002 Elsevier Science NL
Surgical management of aortopulmonary window: a 40-year experience
C.L. Backera,b*,
C. Mavroudisa,b
a Division of Cardiovascular-Thoracic Surgery, M/C #22, Children's Memorial Hospital, 2300 Children's Plaza, Northwestern University Medical School, Chicago, IL, USA
b Department of Surgery, Northwestern University Medical School, Chicago, IL, USA
Received 18 September 2001;
received in revised form 19 December 2001;
accepted 16 January 2002.
* Corresponding author. Tel.: +1-773-880-4378; fax: +1-773-880-3054
e-mail: c-backer{at}northwestern.edu
 |
Abstract
|
|---|
Objectives: An aortopulmonary window (APW) is a communication between the pulmonary artery (PA) and the ascending aorta in the presence of two separate semilunar valves. The purpose of this review is to describe the evolution of surgical techniques and results of surgical correction of APW at a single center over a 40-year time period. Methods: Between 1961 and 2001, 22 patients underwent repair of APW. Age ranged from 11 days to 13 years (median 0.3 years). Associated cardiac lesions included interrupted aortic arch (IAA) (four), right PA origin from the aorta (four), ventricular septal defect (three), atrial septal defect (one), tetralogy of Fallot (one), and transposition of the great arteries (one). Mean preoperative pulmonary vascular resistance was 5.4 U/m2 (n=17). Two patients had attempted ligation without cardiopulmonary bypass (CPB), one patient had division and oversewing of the APW between clamps on CPB. Ten patients had the APW divided on CPB with primary aortic closure. Three patients had circulatory arrest for APW division, IAA repair, and anastomosis right PA to main PA. Most recently, six patients have had open transaortic patch closure (one of these had simultaneous arterial switch, one had simultaneous IAA repair). Follow-up in operative survivors ranges from 1 month to 26 years (median 8 years). Results: There were five early deaths and one late death (pulmonary hypertension) in the first 16 patients where the primary strategy was APW division (37% mortality). There have been no deaths in the most recent six patients having transaortic patch closure. The patients with transaortic patch closure at a maximum of 8 years follow-up are demonstrating normal PA and aortic growth. Conclusions: Early correction of APW with a transaortic patch and repair of all other associated cardiac anomalies at the time of diagnosis is advised.
Key Words: Aorto-pulmonary window • Trans-aortic patch closure
|
1. Introduction
|
|---|
An aortopulmonary window (APW) is a communication between the main pulmonary artery (PA) and the ascending aorta in the presence of two separate semilunar valves. APW is a very uncommon congenital anomaly and there are very few surgical series of more than 20 patients in the literature. Several different classification schemes have been proposed [1–3]. Most recently, Jacobs et al. [4] recommended using the terms proximal, distal, total, and intermediate (Fig. 1
).
View larger version (77K):
[in this window]
[in a new window]
|
Fig. 1. Classification scheme recommended by the Society of Thoracic Surgeons Congenital Heart Surgery Database Committee for aortopulmonary window. Type I is a proximal APW located just above the sinus of Valsalva, a few millimeters above the semilunar valve. Proximal defects are noted to have little inferior rim separating the APW from the semilunar valves. Type II is a distal APW located in the uppermost portion of the ascending aorta. This would correspond to the Richardson type 2 lesion, where the defect overlies a portion of the right PA. Distal defects are noted to have a well-formed inferior rim but little superior rim. Type III is a total defect involving the majority of the ascending aorta. Type IV is the intermediate defect. These defects have adequate superior and inferior rims and are the group most suitable for possible device closure.
|
|
The surgical treatment of APW represents a continuous evolution of techniques that reflect the advances in pediatric cardiac surgery over time. In 1952, Gross [5] reported the successful ligation of an APW through a left thoracotomy. Scott and Sabiston [6] described the method of APW division and suture between clamps in 1953. Repair of an APW using cardiopulmonary bypass (CPB) was reported by Cooley [7] in 1957. He divided the APW between clamps while the patient was on CPB. Morrow et al. [8] reported using CPB with suture of the aortic portion of the APW with a clamp and suturing the PA ‘open’ without a clamp in 1962. The first transaortic closure was reported by Wright et al. [9] in 1968. The first transaortic closure with a patch was reported by Deverall et al. [10] from Great Ormond Street in 1969.
The purpose of this review was to describe the evolution of diagnostic techniques, preoperative management, and surgical techniques at Children's Memorial Hospital since the first procedure was performed 40 years ago. We also review the outcome of the different techniques of surgical correction and describe our current diagnostic and therapeutic approach.
|
2. Patients and methods
|
|---|
Between 1961 and 2001, 22 patients have undergone repair of APW at Children's Memorial Hospital. Age ranged from 11 days to 13 years; median age was 0.3 years and mean age was 1.8±3.4 years, 16 patients were under 1 year of age. Associated cardiac defects are shown in Table 1. No patients in our series had anomalous origin of a coronary artery from the PA. The operative techniques employed evolved considerably over time. They included ligation without the use of CPB (two patients), ligation and division between clamps using CPB (one patient), APW division with CPB with a clamp on the aortic side only (one patient), APW division and primary aortic and PA closure on CPB (nine patients), and most recently, transaortic patch closure (six patients). Three patients had APW division, interrupted aortic arch (IAA) repair, and anastomosis of the right PA to the main PA. Preoperative cardiac catheterization was performed in the first 16 patients and in the patient who had complete transposition (total, 17 patients). The mean preoperative pulmonary vascular resistance was 5.4 Woods units/m2. The most recent six patients (except the patient with transposition) have been operated on the basis of echocardiographic study only. The follow-up in operative survivors ranges from 1 month to 26 years (mean 8.1±6.5 years, median 8 years). The clinical characteristics, surgical techniques, and patient outcomes are summarized in Table 2.
View this table:
[in this window]
[in a new window]
|
Table 2. Clinical characteristics, surgical techniques, and outcomes of patients with aortopulmonary window operated on at Children's Memorial Hospital, 1961–2001a
|
|
2.1. Surgical technique – APW division
CPB is initiated with a single venous cannula and femoral (prior to 1970) or distal aortic cannula. The patient was cooled to 18°C. The ‘shunt’ in initial cases was controlled with digital pressure. The aorta was cross-clamped above the APW and the APW was divided. Cardioplegia was first used in 1977. The aortic side was closed primarily with a two-layer closure. The PA side was closed with a one-layer closure. In the early cases the APW itself was clamped and divided. Initially clamps were placed on both the aorta and PA (like a ductus division), and later the clamp was placed only on the aortic side.
2.2. Surgical technique – transaortic patch closure
CPB is initiated with a single venous cannula and an aortic cannula placed distally on the aortic arch. Upon starting CPB, the right and left main pulmonary arteries are occluded with snares. The patient is cooled to 28°C and a vent is placed in the right superior pulmonary vein. The aorta is cross-clamped and cold blood cardioplegia injected into the aortic root. The aorta is opened with an anterior vertical incision oriented between the non-coronary sinus and the aortic cross-clamp (Fig. 2 ). The APW is identified posteriorly, as are the coronary artery orifices, the cusps of the aortic valve, and the origin and course of both right and left pulmonary arteries. The APW is closed with a 0.4-mm thick Gore-Tex cardiovascular patch (W.L. Gore and Associates, Flagstaff, AZ, USA) anchored with running polypropylene suture. The suturing technique used places all the knots in the suture on the outside of the vessels. The anterior aortotomy is closed with a running polypropylene suture. The heart is de-aired and the patient warmed and weaned from CPB.
View larger version (91K):
[in this window]
[in a new window]
|
Fig. 2. Operative closure of a type III total defect. Not illustrated are the aortic cross-clamp and CPB cannulas. The ascending aorta has been opened with a vertical aortotomy extending from the base of the innominate artery to a point between the non-coronary and right coronary cusps. This aortotomy helps avoid the orifice of the right coronary artery. The defect has been closed with an oval-shaped Gore-Tex cardiovascular patch that has been anchored with a running polypropylene suture. The final suture placement will be such that the knot is tied outside the circulation.
|
|
2.3. Simultaneous repair of APW, IAA, and aortic origin of the PA
The constellation of distal APW, aortic origin of the right PA, intact ventricular septum, patent ductus arteriosus, and IAA was originally described by Berry et al. [11] from Children's Memorial Hospital. It is interesting that all three patients in our series with IAA and right PA from the aorta had a type ‘A’ IAA. Also, none of these patients had diGeorge syndrome. All patients were repaired using circulatory arrest. We have previously reported two of these patients [11]. The first patient was a 32-day-old 2.5-kg child who underwent 8-mm Dacron graft repair of IAA. Right PA-to-main PA continuity was established by tunneling the main PA to the RPA through the APW. This child died of low cardiac output and bleeding several hours postoperatively. The second patient was a 26-day-old 3-kg baby who had the left subclavian artery anastomosed to the descending aorta, division and primary aortic closure of the APW, and a direct anastomosis of the right PA to the main PA. That child required reoperation (left thoracotomy) for repair of arch stenosis (resection with end-to-end anastomosis) at 6 months of age. The third patient was a 30-day-old who underwent end-to-side anastomosis of the descending aorta to the transverse arch, closure of the APW using a partial occlusion clamp on the aorta, with primary aortic closure, and direct anastomosis of the right PA to the resultant orifice in the main PA. The child had preoperative hepatic failure (bilrubin over 20 u/dl). The child died of renal and hepatic failure and sepsis on postoperative day 16. The most recent patient to present with APW and IAA underwent our current approach with a Gore-Tex patch to repair the APW and extended end-to-end anastomosis to repair the IAA. That child required reoperation for homograft aortic arch augmentation 1 year after the initial procedure. Patient #21 in this series had a right PA origin from the ascending aorta but no IAA. That patient had Gore-Tex patch closure of the aorta through an anterior aortotomy and reimplantation of the right PA into the main PA.
2.4. Simultaneous repair of APW and arterial switch operation
This child (patient #19) was referred at the age of 3 years from overseas. He had complete transposition of the great arteries with anteroposterior great vessel relationship. Cardiac catheterization demonstrated a pulmonary vascular resistance of 2.8 U/m2. The APW was quite large (3 cm in diameter). At surgery, the neoaorta was augmented with an aortic homograft patch to fill in the moiety due to the APW. The maneuver of Lecompte was not performed, instead the neopulmonary artery was reconstructed with a 22-mm aortic homograft tube graft, keeping the neopulmonary artery to the left of the aorta and in an anatomic configuration. The coronary artery anatomy was sinus 1, left anterior descending and circumflex arteries, and sinus 2, right coronary artery. There was a persistent left superior vena cava.
|
3. Results
|
|---|
There were five early deaths and one late death in the first 16 patients (37% mortality) where the main strategy was APW division and primary closure. There have been no deaths in the most recent six patients having transaortic patch closure. The early deaths included both patients who had attempted ligation without CPB and one patient who had attempted division and oversewing between clamps on CPB. The first patient in the series had attempted ligation through a left thoracotomy at the age of 2 years in 1961. That patient died the first postoperative day of apnea and bradycardia. Autopsy showed that the APW had been reduced in size from 2 cm in diameter to 0.5 cm in diameter. Autopsy also showed severe pulmonary hypertensive changes. The second death in the series occurred during reoperation 2 months after attempted ligation of the APW without bypass. During reoperation CPB was used and the APW was divided and oversewn. The child died of bleeding and low cardiac output on postoperative day 1. The third death in the series occurred in a neonate that underwent emergent operation at age 11 days for unrelenting acidosis. In retrospect the acidosis was secondary to undiagnosed necrotizing enterocolitis. The child had ventricular fibrillation after the sternotomy and was placed on CPB while receiving CPR. Initial attempt at APW closure was to apply ductus clamps on the aortic and pulmonary sides and divide the fistula. However, there was a tear on the aortic clamp side and the remainder of the repair was performed with circulatory arrest. The child had significant bleeding and expired shortly after surgery. The fourth and fifth early deaths have been described in Section 2.3. There was one late death from pulmonary hypertension 26 years postoperatively. That patient was operated on at the age of 3 years and also had a ventricular septal defect and an atrial septal defect, and a PVR of 11 U/m2. Lung biopsy 9 years postoperatively showed Heath Edward grade III–IV changes. There have been no deaths in the most recent six patients who had transaortic patch closure including one patient with IAA and one patient with transposition of the great arteries. In the patients who have had transaortic patch closure, normal growths of the aorta and PA have been seen at 8 years follow-up by echocardiography. These patients do not have pulmonary hypertension by echocardiographic analysis.
|
4. Discussion
|
|---|
The evolution of the diagnostic and surgical approaches to APW mirrors the changes that have occurred in the field of pediatric cardiac surgery over the same time period. The initial diagnostic approach to APW was with cardiac catheterization. Current diagnostic techniques emphasize non-invasive echocardiographic assessment and in most patients there is no longer a need for cardiac catheterization. The surgical approach began with ligation without the use of CPB, progressed to division between clamps with CPB, progressed to division and primary closure with CPB and aortic cross clamp, and now to transaortic patch closure. The age at time of repair has dropped to the point where almost all children are now repaired as neonates. This has improved the results with regard to complications from pulmonary hypertension. Consistent with the evolution in treatment of other cardiac lesions, transcatheter closure has now also been reported, particularly for intermediate type defects [12].
Mori et al. [1] classified APW into three types, proximal, distal, and total. In the Richardson classification [2], type I is proximal, type II is distal with extension of the communication into the origin of the right PA, and type III is anomalous origin of the right PA from the ascending aorta. Ho et al. [3] modified the Mori classification based on the recent utilization of transcatheter devices to close APW. They included an additional description of an intermediate defect as a defect with adequate superior and inferior rims. Jacobs and the Congenital Heart Surgery Database Committee [4] recently recommended use of hierarchical levels for APW classification. Level 1 separates out APW and PA origin from the ascending aorta. Level 2 separates out APW+IAA, APW types 1–3 (proximal, distal, total), and APW, intermediate type. The different types of defects are shown in Fig. 1. Simple APW has been defined as without any significant associated anomalies, or anomalies requiring minor or simple repair (patent ductus arteriosus, atrial septal defect, patent foramen ovale). Complex APW is a defect occurring with more complex associated anomalies such as IAA, transposition, tetralogy of Fallot, or anomalous origin of the coronary arteries.
Surgical closure of an APW was first reported by Gross [5] in 1952. The patient was a 4-year-old girl he operated on in 1948. Preoperative diagnosis was patent ductus arteriosus, but no open ductus was found. Attention was directed to opening the pericardium and discovering a communication between the ascending aorta and the main PA. The APW was encircled with a ‘1-cm wide linen tape’. The tape was slowly ligated, but suddenly there was some relatively rapid bleeding from the area of the dissection. The tape was then ‘rapidly and tightly drawn down – fortunately all of the bleeding stopped’. In the 1950s and 1960s, there was a progression of surgical closure from dividing and oversewing between clamps [6] to using CPB [7,8].
When Wright and coworkers performed the first transaortic closure in 1968, they noted that most APWs have virtually no length, are usually large, thin-walled, and under high pressure. In addition, they are in an anatomic location where adequate access is difficult to achieve. The transaortic route allows good visualization and preservation of the coronary orifices, the aortic valve leaflets, and the right and left PA orifices. The transaortic approach precludes the necessity for any dissection of the fistula at all with the attendant risks of entering the fistula and having hemorrhage. In at least three patients in our series, intraoperative bleeding following APW division was a significant contributor to early mortality.
The first use of a patch placed using the transaortic route was reported by Deverall and associates [10] of Great Ormond Street in 1969. In the first patient in which a patch was used, the APW was initially closed through a vertical pulmonary arteriotomy. When the aortic cross-clamp was released, they noted that the defect had not been completely closed and there was narrowing of the aorta as well. A vertical aortotomy was then performed that gave a good view of the defect, which was closed with a Dacron patch. The next patient in their series they electively began with a vertical aortotomy and closed it with a Dacron patch. Of note in this particular patient the left coronary orifice was 2 mm away from the lower margin of the defect. The first series of patients (five patients) having a Dacron patch closure through a transaortic approach was reported by Clarke and Richardson [13] in 1976. They noted that the transaortic approach is easy to close, allows the origin of the coronary arteries to be accurately visualized, and allows right PA reconstruction if necessary. The patch placed from the aortic side does not distort either the aorta of PA and is naturally self-sealing.
A summary of the outcomes of patients from different surgical series is shown in Table 3. In the current era, most centers now recommend transaortic patch closure for most patients with an APW [15,17,19]. Most of the morbidity and mortality in these patients are related to the associated cardiac defects, in particular the association of IAA. The late mortality early in our series was related to pulmonary hypertension because many children were not operated on until later in life. The mean age of the first six patients in this series was 4.9 years, the mean age in the last six patients was 0.7 years. Now, all patients are operated on as neonates, before they can develop irreversible pulmonary hypertension. Alternative techniques currently used in some centers include division and patch closure using a ‘sandwich’ technique [20] and division using the PA to patch the aorta, reconstructing the PA with pericardium [21,22]. The ‘sandwich’ technique has been used with success at Boston Children's Hospital [23]. Matsuki et al. [21] reported an 8-day old baby where the anterior wall of the PA was used as a large flap to reconstruct the posterolateral aortic wall. van Son et al. [22] reported using a similar technique successfully in 5- and 9-week-old patients. The difficulty with these techniques is the increased risk of bleeding and the potential for PA or aortic distortion and/or stenosis.
This review of our experience with APW at the Children's Memorial Hospital during a 40-year time period has shown considerable progress in both diagnostic and surgical techniques. The high mortality in the first part of the series was related to a lack of knowledge of the natural history of APW and the technical difficulties of dividing an APW between clamps, with or without the use of CPB. The advent of echocardiography and early surgical intervention, along with the transaortic approach, has led to optimal early and long-term results. Patch closure is associated with minimal difficulties with intraoperative hemorrhage and appears to allow for adequate growth of both the aorta and PA structures. It helps avoid injury to the coronary arteries, aortic valve leaflets, and PA orifices. Complications of pulmonary hypertension are avoided by neonatal repair. Our current strategy for management of infants with APW is echocardiographic diagnosis followed by transaortic patch closure at the time of diagnosis.
|
Footnotes
|
|---|
Presented at the joint 15th Annual Meeting of the European Association for Cardio-thoracic Surgery and the 9th Annual Meeting of the European Society of Thoracic Surgeons, Lisbon, Portugal, September 16–19, 2001.
|
References
|
|---|
-
Mori K., Ando M., Takao A., Ishikawa S., Imai Y. Distal type of aortopulmonary window. Report of 4 cases. Br Heart J 1978;40:681-689.[Abstract/Free Full Text]
-
Richardson J.V., Doty J.B., Rossi N.P., Ehrenhaft J.L. The spectrum of anomalies of aortopulmonary septation. J Thorac Cardiovasc Surg 1979;78:21-27.[Abstract]
-
Ho S.Y., Gerlis L.M., Anderson C., Devine W.A., Smith A. The morphology of aortopulmonary window with regard to their classification and morphogenesis. Cardiol Young 1994;4:146-155.
-
Jacobs J.P., Quintessenza J.A., Gaynor J.W., Burke R.P., Mavroudis C. Congenital Heart Surgery Nomenclature and Database Project: aortopulmonary window. Ann Thorac Surg 2000;69(Suppl):S44-S49.[Abstract/Free Full Text]
-
Gross R.E. Surgical closure of an aortic septal defect. Circulation 1952;5:858-863.[Medline]
-
Scott H.W., Jr, Sabiston D.C., Jr Surgical treatment for congenital aorticopulmonary fistula: experimental and clinical aspects. J Thorac Surg 1953;25:26.
-
Cooley D.A., McNamara D.G., Latson J.R. Aorticopulmonary septal defect: diagnosis and surgical treatment. Surgery 1957;42:101-120.
-
Morrow A.G., Greenfield L.J., Braunwald E. Congenital aortopulmonary septal defect: clinical and hemodynamic findings, surgical technic and results of operative correction. Circulation 1962;25:463-476.[Abstract/Free Full Text]
-
Wright J.S., Freeman R., Johnston J.B. Aorto-pulmonary fenestration. A technique of surgical management. J Thorac Cardiovasc Surg 1968;55:280-283.[Medline]
-
Deverall P.B., Lincoln J.C., Aberdeen E., Bonham-Carter R.E., Waterston D.J. Aortopulmonary window. J Thorac Cardiovasc Surg 1969;57:479-486.[Medline]
-
Berry T.E., Bharati S., Muster A.J., Idriss F.S., Santucci B., Lev M., Paul M.H. Distal aortopulmonary septal defect, aortic origin of the right pulmonary artery, intact ventricular septum, patent ductus arteriosus and hypoplasia of the aortic isthmus: a newly recognized syndrome. Am J Cardiol 1982;49:108-116.[Medline]
-
Stamato T., Benson L.N., Smallhorn J.F., Freedom R.M. Transcatheter closure of an aortopulmonary window with a modified double umbrella occluder system. Catheter Cardiovasc Diagn 1995;35:165-167.[Medline]
-
Clarke C.P., Richardson J.P. The management of aortopulmonary window: advantages of transaortic closure with a Dacron patch. J Thorac Cardiovasc Surg 1976;72:48-51.[Abstract]
-
Blieden L.C., Moller J.H. Aorticopulmonary septal defect. An experience with 17 patients. Br Heart J 1974;36:630-635.[Free Full Text]
-
Doty D.B., Richardson J.V., Falkovsky G.E., Gordonova M.I., Burakovsky V.I. Aortopulmonary septal defect: hemodynamics, angiography, and operation. Ann Thorac Surg 1981;32:244-250.[Abstract]
-
Tiraboschi R., Salomone G., Crupi G., Manasse E., Salim A., Carminati M., Vanini V., Parenzan L. Aortpulmonary window in the first year of life: report on 11 surgical cases. Ann Thorac Surg 1988;46:438-441.[Abstract]
-
van Son J.A.M., Puga F.J., Danielson G.K., Seward J.B., Mair D.D., Schaff H.V., Ilstrup D.M. Aortopulmonary window: factors associated with early and late success after surgical treatment. Mayo Clin Proc 1993;68:128-133.[Medline]
-
Tkebuchava T., von Segesser L.K., Vogt P.R., Bauersfeld U., Jenni R., Kunzli A., Lachat M., Turina M. Congenital aortopulmonary window: diagnosis, surgical technique and long-term results. Eur J Cardiothorac Surg 1997;11:293-297.[Abstract]
-
McElhinney D.B., Reddy V.M., Tworetzky W., Silverman N.H., Hanley F.L. Early and late results after repair of aortopulmonary septal defect and associated anomalies in infants <6 months of age. Am J Cardiol 1998;81:195-201.[Medline]
-
Johansson L., Michaelsson M., Westerholm C.J., Aberg T. Aortopulmonary window: a new operative approach. Ann Thorac Surg 1978;25:564-567.[Medline]
-
Matsuki O., Yagihara T., Yamamoto F., Nishigaki K., Uemura H., Kawashima Y. New surgical technique for total-defect aortopulmonary window. Ann Thorac Surg 1992;54:991-992.[Abstract]
-
van Son J.A., Hambsch J., Mohr F.W. Anatomical reconstruction of aorta and pulmonary trunk in patients with an aortopulmonary window. Ann Thorac Surg 2000;70:674-675.[Abstract/Free Full Text]
-
Castañeda A., Jonas R., Mayer J., Hanley F. Aortopulmonary window. Cardiac surgery of the neonate and infant. Philadelphia, PA: W.B. Saunders Co, 1994.
This article has been cited by other articles:
|
|
|
|
 
C. R. Patel, G. L. Smith, and J. R. Lane
Tetralogy of Fallot With an Aortopulmonary Septal Defect: Prenatal Diagnosis
J. Ultrasound Med.,
August 1, 2007;
26(8):
1115 - 1117.
[Full Text]
[PDF]
|
|
|
|
|
|
|
A. Bhan, M. Gupta, S. Abraham, R. Sharma, S. S. Kothari, and R. Juneja
Surgical experience of aortopulmonary window repair in infants
Interactive CardioVascular and Thoracic Surgery,
April 1, 2007;
6(2):
200 - 203.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
K. Adluri, D. J. Barron, and W. J. Brawn
D-Transposition of the Great Arteries With an Aortopulmonary Window: A New Corrective Technique
Ann. Thorac. Surg.,
March 1, 2005;
79(3):
1066 - 1067.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
1. Introduction
