| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Eur J Cardiothorac Surg 2002;22:192-199
© 2002 Elsevier Science NL
a Department of Paediatric Cardiac Surgery, Ospedale Pediatrico Apuano ‘G. Pasquinucci’, Via Aurelia Sud, 54100 Massa, Italy
b Department of Paediatric Cardiology, Ospedale Pediatrico Apuano ‘G. Pasquinucci’, Via Aurelia Sud, 54100 Massa, Italy
Received 20 August 2001; received in revised form 9 April 2002; accepted 22 April 2002.
* Corresponding author. Tel.: +39-585-493522; fax: +39-585-493616
e-mail: acerillo{at}yahoo.com
 |
Abstract |
|---|
 Top Abstract 1. Introduction2. Materials and methods3. Results4. Discussion5. ConclusionReferences |
|---|
Key Words: Single ventricle • Pulmonary artery banding • VSD enlargement
|
1. Introduction |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. Discussion5. ConclusionReferences |
|---|
Subaortic stenosis may be present at birth or develop after palliative procedures [4,6]. In particular, pulmonary artery banding has been indicated as to predispose to the development of a restrictive VSD. By increasing afterload, pulmonary artery banding causes hypertrophy of the dominant ventricle [4]. It also acutely reduces the QP/QS ratio, thus altering ventricular geometry [6], reducing ventricular volume, and increasing ventricular wall thickness. Once established, subaortic stenosis brings about myocardial hypertrophy, increases myocardial wall thickness and reduces ventricular compliance, and these alterations constitute known risk factors for the outcome of a one-ventricle repair [7].
Since the first reports of the potentially dangerous effect of pulmonary artery banding, many alternative surgical procedures have been proposed [7–13]. Palliative arterial switch operation, Damus–Kaye–Stansel connection (DKS), and Norwood-like procedures, all enable to protect pulmonary vascular bed and to bypass the possible obstruction site represented by the ventricular septal defect.
The purpose of this study was to analyse the impact of preoperative and procedural characteristics, including pulmonary artery banding, that could increase the risk of developing a restrictive ventricular septal defect. We also report our operative and long-term results with a staged approach based on initial pulmonary artery banding and, eventually, treatment of the restrictive VSD by direct muscular resection.
|
2. Materials and methods |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. Discussion5. ConclusionReferences |
|---|
Twenty-four patients were finally included in the analysis (Table 1). There were 12 females (50%) and 12 males. Median age at operation was 7.5 days (range: 1–376). Median weight was 3.5 kg (range: 1.9–6.3). Thirteen (54.7%) patients had double inlet left ventricle and transposed great arteries (DILV-TGA). Seven patients (29.7%) had absent right atrioventricular connection and transposed great arteries (tricuspid atresia and transposed great arteries, TA-TGA), and four (16.7%) had absent left atrioventricular connection and concordant ventriculo-arterial connection (mitral atresia without aortic atresia, MA). Associated anomalies were present in 15 patients (Table 1). Five patients (20.83%) had coarctation of the aorta (DILV-TGA: 1) or aortic hypoplasia (MA: 2; DILV-TGA: 1; TA-TGA: 1). Three patients (DILV-TGA: 2; TA-TGA: 1) had mild pulmonary stenosis, and five (TA-TGA: 3; DILV-TGA: 2) had severe pulmonary stenosis or pulmonary atresia.
|
Subaortic stenosis was present at diagnosis in three patients (12.5%): one with TA-TGA and aortic hypoplasia, in which the VSD was restrictive at birth, one with DILV-TGA, infracardiac total anomalous pulmonary venous connection (TAPVC) and a restrictive VSD, and a third with MA and a subaortic fibromuscular ring (Table 1).
2.3. Operative management
Sixteen patients (66.66%; DILV-TGA: 10; MA: 4; TA-TGA: 2) with pulmonary overflow were initially treated by pulmonary artery banding (Fig. 1)
.
|
Six patients undergoing PAB required additional surgical procedures for the presence of associated anomalies (Table 1). One with MA and a subaortic fibromuscular ring also underwent subaortic fibromuscular ring resection through the aorta, during cardiopulmonary bypass and cardioplegic arrest. One with DILV-TGA and coarctation of the aorta underwent coarctation repair by end to end anastomosis. This is the only patient in our series in which PAB was performed through a left thoracotomy. Three patients (DILV-TGA: 1; MA: 2) with associated aortic hypoplasia also underwent aortic enlargement with autologous pericardium during total circulatory arrest and deep hypothermia.
A patient with DILV-TGA, restrictive VSD and TAPVC underwent correction of TAPVC and VSD enlargement through the atriotomy, during cardiopulmonary bypass and cardioplegic arrest. On rewarming, the pulmonary artery was banded as described above. This is the only patient in our series in which direct VSD enlargement was performed during neonatal life. A second patient with TA-TGA, restrictive VSD and aortic hypoplasia (Fig. 1 and Table 1) underwent Damus–Kaye–Stansel connection and aortic enlargement with autologous pericardium (see below).
Five patients (20.83%; TA-TGA: 3; DILV-TGA: 2) initially required a modified Blalock–Taussig shunt for the presence of pulmonary stenosis or atresia (Table 1 and Fig. 2) . This was performed through a median sternotomy, with a 4- or 5-mm polytetrafluoroethylene graft anastomosed to the right subclavian artery proximally and to the superior aspect of the right pulmonary artery distally with two continuous 6/0 polypropylene sutures.
|
Bi-directional Glenn shunt was performed during cardiopulmonary bypass with a beating heart. Selective venous cannulation of the inferior vena cava and of the innominate vein achieved venous drainage. The superior vena cava was transected at his junction with the right atrium and anastomosed to the superior aspect of the right pulmonary artery with a continuous 6/0 polypropylene suture in the posterior and lateral aspects and with simple interrupted 6/0 polypropylene stitches in the anterior aspect.
2.4. Definitive repair
The presence of an incomplete ventricle is considered indication to a one-ventricle repair. This was obtained by a staged approach involving a bi-directional Glenn shunt, performed at around 6 months of age and subsequent completion of the operation, at 2–4 years of age. Total cavopulmonary connection was performed during cardiopulmonary bypass with a beating heart. The inferior vena cava was transected and anastomosed to a 16–22-mm Gore-Tex conduit with a continuous 5/0 Gore-Tex suture. This was then anastomosed to the inferior aspect of the right pulmonary artery with a continuous 5/0 Gore-Tex suture.
2.5. Management of subaortic stenosis (Table 1)
In two patients with MA (Table 1), subaortic stenosis was due to the presence of fibromuscular tissue along the systemic outflow tract. They were treated by fibromuscular tissue resection trough the aorta, the first in the neonatal period and the second at Glenn.
Subaortic stenosis due to a restrictive VSD was preferably managed by direct VSD enlargement, performed during cardiopulmonary bypass with cardioplegic arrest, either through the atrium (five cases) or through the aorta (two cases). In order to avoid the conduction tissue, the VSD was enlarged in a superior and anterior direction, as described in the literature [3,7]. One patient with DILV-TGA, restrictive VSD and TAPVC (see above) underwent VSD enlargement through the atrium during the neonatal era. Despite the small size of the heart, the VSD was easily visualized through the right-sided AV valve when the right atrium was opened for TAPVC correction.
A Damus–Kaye–Stansel connection was employed when VSD enlargement was considered not feasible through the atrium or the aorta. In our series, a DKS was performed in two cases. The first one was a neonate with TA-TGA and aortic hypoplasia that also underwent aortic enlargement. The second was an infant with TA-TGA and banded pulmonary artery, that underwent bi-directional Glenn plus DKS. In both cases the VSD was not easily accessible from the aorta or the atrium. DKS was performed during cardiopulmonary bypass with cardioplegic arrest. After excision of a portion of the proximal aortic wall, the pulmonary artery was transected and end-to-side anastomosed to the aorta with a continuous 5/0 polypropylene suture.
2.6. Data
Considered variables were obtained by a retrospective review of clinical records, outpatient records, and operative registry. Continuous variables were analysed by the Mann–Whitney U-test. Dichotomous variables were analysed by Fisher's exact test. Overall survival was estimated by the Kaplan–Meier method and comparison between unadjusted overall group survival relative to baseline characteristics was assessed by a log-rank test. Long-term freedom from subaortic obstruction was evaluated by the Kaplan–Meier method. Significance of considered variables was assessed by a log-rank test. P
0.05 was considered significant.
|
3. Results |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. Discussion5. ConclusionReferences |
|---|
|
3.2. Subsequent operations (Figs. 1, 2 and Table 1)
Of the 21 survived patients, 20 have received a bi-directional Glenn anastomosis, either as a first procedure (2) or as the second one (18). One of them, with DILV-TGA and severe pulmonary stenosis requiring a shunt in the neonatal age, died early postoperatively (cumulative mortality: 16.67%).
Nineteen patients survived after bi-directional Glenn shunt. Twelve of these subsequently underwent total cavopulmonary connection, with no operative deaths. One patient with DILV-TGA and no associated anomalies already treated by pulmonary artery banding and bi-directional Glenn, required Fontan take-down for the appearance, early postoperatively, of a severe low output syndrome.
3.3. Follow-up
Overall, four patients (16.67%) died during the study period. All 20 surviving patients are periodically seen at our outpatients’ service, and actually are asymptomatic or paucisymptomatic (Table 1). Eleven patients (45.83%) have achieved a complete repair. One additional patient required Fontan take-down, and up to now has been doing well. Eight patients are awaiting a definitive repair.
Five-year cumulative survival (Kaplan–Meier) was 85.7% (95% confidence limit (CL): 70.7–100%) (Fig. 3) . The only variable associated with long-term outcome was the presence of coarctation or hypoplasia of the aorta (Table 3).
|
|
|
|
Among the eight patients with a restrictive VSD, two received a Damus–Kaye–Stansel connection. The first one was a neonate with TA-TGA and aortic hypoplasia that also underwent aortic arch enlargement. The second was an infant with TA-TGA and banded pulmonary artery, that underwent bi-directional Glenn plus DKS. As already stated, in this patient the VSD was partially covered by accessory tissue from the left-sided AV valve, and VSD enlargement was judged not feasible.
Six patients underwent VSD enlargement, in the neonatal period (1), at Glenn (1), or in concomitance with total cavopulmonary connection (4). One of the patients treated at total cavopulmonary connection presented at postoperative cardiac catheterization with an isoproterenol provoked gradient of about 10 mmHg, in the absence of a basal gradient, and received no further treatment. A second patient had a basal gradient of about 20 mmHg and underwent VSD re-enlargement through an atriotomy, with no complications. None of the seven procedures performed was complicated by the occurrence of complete heart block.
Cumulative freedom from subaortic stenosis (Kaplan–Meier) at 6 months, 2 and 5 years was 83.3% (95% CL: 68.4–98.2%), 74.8% (95% CL: 57.3–92.3%), and 54.0% (95% CL: 32.5–75.5%), respectively (Fig. 4).
None of the tested variables, including previous pulmonary artery banding, was significantly associated with the development of subaortic stenosis (Table 4).
Five patients with a previous history of subaortic stenosis have achieved a definitive repair, including one that subsequently underwent Fontan takedown. Four are awaiting the completion of total cavopulmonary connection.
|
4. Discussion |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. Discussion5. ConclusionReferences |
|---|
The relationships between pulmonary artery banding, myocardial hypertrophy and the appearance of a restrictive VSD were clearly explained by the group of Toronto [2,4]. By increasing the dominant ventricle afterload, pulmonary artery banding provokes concentric myocardial hypertrophy, increased myocardial mass, reduced ventricular compliance and finally ventricular diastolic dysfunction [4,11]. Moreover, it acutely alters QP/QS ratio, thus reducing volume overload of the dominant ventricle, altering ventricular geometry and increasing myocardial wall thickness [6,11]. These two factors have been indicated as to predispose to the development of a restrictive VSD.
In our series, neonatal pulmonary artery banding was not associated with an increased risk of developing a restrictive VSD, nor with an increased risk of long term failure of the dominant ventricle. In our opinion, there are three possible explanations for this. The first one resides in the nature of patients that actually did not receive pulmonary artery banding as a first operation. Apart from the case of a patient with TA-TGA, hypoplastic aorta and restrictive VSD that underwent neonatal Damus–Kaye–Stansel, all the others had pulmonary stenosis severe enough to protect the pulmonary vascular bed, an associated lesion that we encountered in a surprising number of cases [1]. As pulmonary stenosis could itself have acted as a pulmonary band, this may have biased our statistics. The second reason probably relates to the time interval that occurred between pulmonary artery banding and bi-directional Glenn shunt. We use to perform this operation around the age of six months, leaving a time interval possibly too short to allow the appearance of myocardial hypertrophy severe enough to restrict the VSD. Careful monitoring of patient conditions and repeated echocardiograms should suffice to prevent the appearance of this complication. Finally, it is reasonable to suppose that factors other than myocardial hypertrophy also play a role in the genesis of the restrictive VSD. It is noteworthy that in 50% of our patients that developed a restrictive VSD, it appeared after a bi-directional Glenn shunt. Rychik and co-workers [6] clearly demonstrated the role of ventricular volume and geometry alterations carried out by volume reducing surgery in determining the appearance of subaortic obstruction. Our data seem to support the idea that volume reduction may actually play a major role in the genesis of the restrictive VSD.
At present, the best therapeutic solution for patient with univentricular atrioventricular connection is represented by the total cavopulmonary connection. Diastolic dysfunction secondary to myocardial hypertrophy, and pulmonary vascular disease caused by long-standing pulmonary overflow are two among the most powerful risk factors for late failure of a one-ventricle repair. For this reason, preparative palliative procedures should be able to protect the pulmonary vascular bed without damaging the dominant ventricle.
The rationale for palliative arterial switch operation [7,9] is to protect the pulmonary circulation by translocating the pulmonary artery over the obstructed incomplete ventricle outflow tract, creating in the mean time an unobstructed path from the dominant ventricle to aorta. This approach has been largely abandoned, mostly because of the scarce reliability of the VSD in regulating pulmonary blood flow, resulting either in pulmonary overflow or in cyanosis.
Damus–Kaye–Stansel connection [10,12] or, in the presence of an hypoplastic aorta, Norwood-like procedures [13,14], allow to bypass the obstruction site represented by the VSD. In recent years tremendous progresses have been reported in the management of hypoplastic left heart syndrome, and in selected institutions the Norwood operation is now performed with an acceptable mortality, making this procedure, at last in theory, the one of choice for patients with an hypoplastic aorta originating from a rudimentary ventricle. Unfortunately, in our hands the Norwood stage I operation still carries a high mortality (to give a figure, operative mortality after stage I Norwood procedure has been 50% in 1997 and 40% in 1999). For this reason at our institution the recourse to this procedure has been limited to classic hypoplastic left heart syndrome. The improved perioperative management of Norwood patients will probably alter this figure in the next years [14].
Recently, good results have been reported with a pulmonary artery banding protocol similar to ours. Jensen and co-workers [15] reported 19% mortality in a series of 26 patients with DILV-TGA or TA-TGA initially treated by pulmonary artery banding, 16 of who subsequently developed subaortic obstruction. Webber and co-workers [16] reported similar results. This approach allows treating neonate patients with a simple, fast, and rather sure procedure, without avoiding the possibility to treat subaortic stenosis should it appear. Amin and co-workers [17] and Deanen et al. [18] have in fact recently demonstrated that pulmonary artery banding does not compromise the pulmonary valve, thus allowing surgeons to perform a Damus–Kaye–Stansel procedure should it became necessary.
Treatment of restrictive VSD by VSD enlargement has been reported to expose patients to the occurrence of various complications, some of which are related to the ventriculotomic access (poor ventricular function, aneurysm formation), while others are explained by surgical damage of the conduction tissue (complete heart block [19]). For this reason subaortic obstruction is now preferably managed by a Damus–Kaye–Stansel connection at many institutions [11]. In our series, VSD enlargement was always accomplished either through the aorta or through the atrium, and was never complicated by the occurrence of complete heart block. This is at least in part due to the fact that the only neonate in our series that required enlargement of the VSD was a girl with DILV-TGA and associated TAPVC, in which the VSD was easily visualized from the atrium at operation. As the transaortic and the transatrial approach are not always feasible in the neonatal era, it is probable that in the neonate with a restrictive VSD other strategies (Damus–Kaye–Stansel or Norwood procedure) are preferable.
Patients with associated coarctation or hypoplasia of the aorta constitute a particularly high-risk subgroup. In our series, three of five such patients died after the first palliation. The coexistence of an obstruction to flow at the aortic and subaortic level should probably be considered as an indication to the Norwood procedure.
Our current strategy for patients with univentricular atrioventricular connection and aorta originating from the incomplete ventricle is based on the following considerations. (1) Protect pulmonary circulation by pulmonary artery banding (in the absence of pulmonary stenosis or atresia). (2) Create a Glenn anastomosis around the sixth month of age, thus avoiding excessive myocardial hypertrophy. During this period, careful echocardiographic monitoring of the VSD is mandatory. (3) Complete the one-ventricle repair by an extracardiac conduit, between the age of 2–4 years. Subaortic stenosis is preferably managed by direct VSD enlargement, associating, when possible, this procedure to one of the surgical steps mentioned above. It is not possible, from our data, to draw conclusions about the best strategy for treatment of the neonate with a restrictive VSD.
|
5. Conclusion |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. Discussion5. ConclusionReferences |
|---|
Neonates with subaortic stenosis constitute a particularly high-risk subgroup, and are probably best managed by different approaches (Damus–Kaye–Stansel or Norwood operation).
|
References |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. Discussion5. ConclusionReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
