| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Pediatric Cardiac Surgery, Medical University of South Carolina, Charleston, South Carolina
b Pediatric Cardiology, Medical University of South Carolina, Charleston, South Carolina
Accepted for publication March 21, 2008.
* Address correspondence to Dr Bradley, Pediatric Cardiac Surgery, Medical University of South Carolina, 96 Jonathan Lucas St., Charleston, SC 29425 (Email: bradlesm{at}musc.edu).
Presented at the Fifty-fourth Annual Meeting of the Southern Thoracic Surgical Association, Bonita Springs, FL, Nov 7–10, 2007.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
Methods: Between August 2004 and July 2007, 10 infants with biventricular hearts and inadequate pulmonary blood flow underwent palliation with an RV-PA shunt. Median age was 9 days (range, 4 to 86), weight was 3.0 kg (1.7 to 4.5), and 4 of 10 patients weighed less than 2.5 kg. Shunts were nonvalved Gore-Tex (W.L. Gore Assoc, Flagstaff, AZ), and size was 6 mm (n = 5) or 5 mm (n = 5).
Results: There were no operative deaths. Median oxygen saturation at hospital discharge was 95% (87 to 98). In 2 patients the shunt was partially narrowed with a metal clip; they underwent successful balloon dilation 6 months after shunt placement. Eight patients have undergone two-ventricle repair 6 to 17 months after shunt placement. At the time of complete repair, oxygen saturation was 86 ± 1% and weight was 7.7 ± 1.7 kg. Repairs included a valved RV-to-PA conduit, 14 to 16 mm in diameter. There was one interstage death.
Conclusions: The RV-PA shunt provides successful palliation in infants with biventricular heart disease and inadequate pulmonary blood flow. It can be used in low birth weight infants and allows significant growth with protection of oxygen saturation prior to complete repair. Partial clipping of the shunt with subsequent balloon dilation is an option to prolong palliation. These results compare favorably with those of a modified Blalock-Taussig shunt or single stage complete repair.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
The right ventricle-to-pulmonary artery (RV-PA) shunt is a potential alternative to the modified Blalock-Taussig shunt. It has had extensive use in patients with hypoplastic left heart syndrome (HLHS). In neonates with HLHS, the RV-PA shunt results in higher diastolic blood pressure, confers a greater resistance to physiologic insults, and may improve outcomes, particularly in low birth weight babies [7–9]. A palliative RV-PA conduit has also been used in patients with biventricular hearts, whose pulmonary blood flow comes from major aortopulmonary collaterals, to encourage the growth of diminutive central pulmonary arteries [10, 11]. The use of an RV-PA shunt as a single, controlled source of pulmonary blood flow has not been previously described in patients with biventricular hearts. The aim of this report is to assess the results of palliation with an RV-PA shunt in infants with biventricular hearts, examining operative and midterm survival, adequacy of palliation, and progression to a successful two-ventricle repair.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
Patient diagnoses were tetralogy of Fallot with pulmonary atresia in 6 patients, tetralogy with atrioventricular septal defect in 2, tetralogy with atrioventricular septal defect and pulmonary atresia in 1, and transposition with ventricular septal defect and pulmonary atresia in 1 (Table 1). The patients with tetralogy of Fallot with pulmonary atresia had confluent central pulmonary arteries supplied by a ductus, and no aortopulmonary collaterals. The 8 patients with pulmonary atresia were dependent on a prostaglandin infusion to maintain ductal patency. Four patients had patterns of aortic arch branching, which were judged to potentially compromise placement of a modified Blalock-Taussig shunt. Three had a right arch with mirror-image head vessels, and a ductus originating from the left innominate artery. One had a left arch with aberrant right subclavian artery.
|
Associated procedures at the time of RV-PA shunt placement were as follows. In the 8 patients with pulmonary atresia, a patent ductus arteriosus, which provided pulmonary blood flow, was divided. In the remaining 2 patients with pulmonary stenosis (patients 4 and 6), the native right ventricular outflow tract was judged to be too small for eventual use and the main pulmonary artery was divided prior to placing the RV-PA shunt. Thus, the RV-PA shunt provided the only source of pulmonary blood flow after operation. Two patients also underwent pulmonary artery reconstruction. After separation from cardiopulmonary bypass, two patients had evidence of pulmonary overcirculation, manifested by low systemic blood pressure, and high systemic oxygen saturation. In each, the shunt was partially narrowed by placement of a metal clip in its midportion, with a plan to lengthen the period of palliation by subsequent balloon dilation of the clipped shunt in the catheterization laboratory.
In an effort to avoid shunt thrombosis, all patients were administered intravenous heparin once operative bleeding had stopped (target partial thromboplastin time 60 to 80 seconds). When enteric intake resumed, heparin was converted to aspirin, which was continued until biventricular repair. The timing of complete repair was generally dictated by the patient's systemic oxygen saturation, rather than targeting a particular patient age or size.
Data Collection/Follow-Up
All records were retrospectively reviewed. All patients were followed within the Children's Heart Program of South Carolina, a pediatric cardiology network providing coordinated congenital cardiac care to the state.
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
Interstage Course
Cardiac catheterization was carried out in 8 patients prior to two-ventricle repair; one patient underwent repair based on computed tomographic angiogram alone. The median time interval from RV-PA shunt placement to catheterization was 6.5 months (4.8 to 9.2 months). The median pulmonary artery pressure at catheterization was 12 mm Hg, with a range of 10 to 15 mm Hg. The 2 patients with a clipped shunt underwent successful balloon dilation of the narrowed shunt, 6 months after shunt placement (Fig 1). This allowed delay of two-ventricle repair for a further 4, and 11 months, respectively. One other patient underwent interstage intervention; cardiac catheterization with placement of a stent in the proximal RV-PA shunt at 6.5 months of age. This patient died suddenly at home at 9 months of age, after an upper respiratory infection. No autopsy was obtained.
|
|
Comment |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
Alternatives to palliation with an RV-PA shunt include primary complete repair or palliation with a different type of shunt. Primary complete repair is now the preferred approach to many heart defects in neonates and infants. However, early complete repair remains a challenge in patients like those in the current series, with anatomic diagnoses of tetralogy of Fallot with pulmonary atresia, tetralogy with atrioventricular septal defect, and transposition with pulmonary atresia. Excellent results have been reported for early primary repair of tetralogy of Fallot [12, 13]. However, within this group of patients, Pigula and colleagues [12] found a diagnosis of tetralogy with pulmonary atresia to be a significant risk factor both for reoperation and for death; early or late mortality occurred in 19% of patients with this diagnosis. Patients with the more complex combination of tetralogy of Fallot with complete atrioventricular septal defect have not generally undergone repair in early infancy. In several previous series, the mean age at repair was 3.5 to 4.5 years, with the youngest patients to undergo complete repair ranging from age 5 months to 1.5 years [14–16]. Primary complete repair of the anatomic defects in the current series of patients would have also involved placement of an RV-PA conduit in a small infant. These considerations weighed against early complete repair in the particular patients in this report.
Despite the popularity and benefits of early complete repair, many infants with biventricular heart disease and inadequate pulmonary blood flow currently undergo initial palliation. Data from the Society of Thoracic Surgeons (STS) and the European Association for Cardiothoracic Surgery (EACTS) congenital databases for the years 2002 to 2003 show that infants with tetralogy of Fallot underwent palliation in 62% (STS) and 70% (EACTS) of cases [17]. At many institutions, initial palliation with a shunt is used in all patients born with tetralogy with pulmonary atresia [18]. In several series of repair for tetralogy with atrioventricular septal defect, initial palliation with a shunt was employed in 55% to 82% of the patients [14–16].
Traditional palliation of such patients has been provided by a modified Blalock-Taussig shunt [1]. However, this approach carries a risk of both operative and interstage mortality. Operative mortality for placement of a modified Blalock-Taussig shunt in the current STS congenital database is 6.7% [2]. Interstage mortality (between hospital discharge and two-ventricle repair) has been reported in 7% to 13% of patients with biventricular anatomy [3, 4]. In a large, multiinstitutional study of systemic-to-pulmonary artery shunts, patients with tetralogy of Fallot had an incidence at one-year postshunt placement of thrombosis 5%, and of death 15% [5]. Successful palliation with a shunt is particularly difficult in low birth weight infants [6]. Abnormalities of aortic arch branching can also make placement of a modified Blalock-Taussig shunt difficult. Three patients in our group had a right aortic arch and mirror-image head vessels with a patent ductus arteriosus originating from the left innominate artery. Placement of a shunt from the innominate artery in these patients could have involved ductal tissue at the proximal shunt site. Another of our patients had a left arch with aberrant right subclavian artery. Other options for a proximal shunt site in these patients would have been the carotid artery or ascending aorta.
Compared with a modified Blalock-Taussig shunt, the RV-PA shunt offers several advantages, both real and potential. It avoids issues with proximal shunt location in patients with abnormal aortic arch branching. Pulmonary blood flow occurs only during cardiac systole; diastolic runoff resulting in lower diastolic and coronary perfusion pressure does not occur. This allows placement of a larger diameter shunt than a modified Blalock-Taussig shunt (generally 5 or 6 mm instead of 3.5 or 4 mm). Most modified Blalock-Taussig shunts develop stenosis due to myofibroblastic proliferation; by the time of takedown, 21% have greater than 50% obstruction [19]. Because smaller shunt size is a risk factor for stenosis, a larger diameter may decrease the incidence of shunt stenosis. Previous use of the RV-PA shunt in patients with hypoplastic left heart syndrome has suggested several other advantages. It may be particularly suited for use in low birth weight babies [8, 9]. It appears to provide greater resistance to physiologic insults such as cardiopulmonary arrest [7]. Finally, it may confer a lower risk of interstage mortality than a modified Blalock-Taussig shunt [20–24].
The RV-PA shunt also has some disadvantages. Unlike the modified Blalock-Taussig shunt, it is placed with the use of cardiopulmonary bypass. While an RV-PA shunt could potentially be placed without aortic cross-clamping, we choose to use a short period of cardiac arrest to facilitate the proximal anastomosis, and avoid air embolism. The RV-PA shunt also has the disadvantage of requiring a ventriculotomy. In the experience with HLHS, it is not clear whether the need for a ventriculotomy is detrimental. Early hemodynamic outcome, including systemic oxygen delivery after a Norwood procedure with use of either an RV-PA or modified Blalock-Taussig shunt, is similar [25, 26]. Two large, concurrent comparisons of the RV-PA and modified Blalock-Taussig shunt have shown similar outcomes through the second stage of palliation [24, 27]. Qualitative systolic ventricular function using the two approaches has been observed to be better [28], the same [29], or worse [27] with an RV-PA shunt. In the current series, we chose to limit use of an RV-PA shunt to patients in whom the subsequent two-ventricle repair was planned to include a right ventriculotomy for placement of an RV-to-PA conduit. Finally, while there were no operative deaths at the time of RV-PA shunt placement or two-ventricle repair, one patient died between operations. This emphasizes the fact that patients palliated with a shunt of any type remain at risk for sudden death and should be closely followed and expeditiously moved to two-ventricle repair.
In our experience, the only anatomic exclusion criterion for use of an RV-PA shunt is the presence of an important coronary artery crossing the right ventricular outflow tract. During the period of this study this occurred in one patient, who had a left anterior descending (coronary artery) originating from the right coronary and was palliated with a modified Blalock-Taussig shunt. This patient subsequently underwent successful repair with an RV-PA conduit.
In conclusion, the RV-PA shunt provides successful palliation in young infants with complex biventricular heart disease and inadequate pulmonary blood flow. It is applicable in low birth weight neonates and patients with abnormal aortic arch branching. It allows significant patient growth with protection of systemic oxygen saturation prior to two-ventricle repair. Partial clipping of the shunt with subsequent balloon dilation is an option to prolong palliation. It may be particularly appropriate in patients who will eventually have a ventriculotomy for placement of an RV-PA conduit. The current results compare favorably with those of other approaches, which include a modified Blalock-Taussig shunt or single-stage complete repair.
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
It is a technique that has been used by others in patients in which the pulmonary arteries need to be rehabilitated. We use RV to PA conduits in patients with tetralogy and pulmonary atresia with small central pulmonary arteries, and it does have the attractiveness of allowing the cardiologists to then have access to these pulmonary arteries. But, as I take it, these are patients that would possibly be repairable; they have a ductus and fairly good pulmonary arteries. I would like to have you respond to three questions so that we might better understand the utility of this operation, which may be very beneficial for selected patients.
Number one, which are the patients that you would choose to palliate as opposed to repair in your institution? And we all are different in our approaches, and so there may be people here who, like you, would prefer to palliate similar patients. Secondly, what are the technical differences between doing the shunt in repairable patients versus the standard Sano, where you are putting the shunt to a Gore-Tex or a homograft patch on fairly large pulmonary arteries? Is there a difference in how you construct this shunt? The final question I have for you is about the novel idea of clipping a shunt. I am wondering if you have extended that into your patients with hypoplastic left heart syndrome to allow you to control shunt flow in your early postoperative period and then improve the shunt flow when the patient is more recovered and has more cardiac output.
It is an interesting idea and it may have some utility and I congratulate you on your creativity.
DR BRADLEY: Thanks a lot for your comments, Ross. First of all, in terms of complete repair, all of these patients technically could have undergone a complete repair as neonates, but I think you have to pick your patients. To answer your question more specifically, in patients with tetralogy of Fallot, pulmonary atresia and adequate central pulmonary arteries supplied by a ductus, if they are a reasonable-sized baby, let's say over 3 kg, and have short segment pulmonary atresia, so that they can be repaired with a transannular patch rather than a conduit and therefore potentially have a one-time operation, then I would do a complete repair as a neonate. There were a few patients during this time period that had that approach taken and were not included in the 10 patients I talked about this morning.
On the other hand, the things that would push me to use this approach would be, in particular, if the patient is small at the time of birth, 2.5 kg or less, and clearly will require eventual repair with an RV to PA conduit. The size of the conduit one can place in a neonate of this size is relatively small, and will bring them back for reoperation potentially as early as an RV-PA shunt. Several of the patients in the current series had tetralogy with AV (atrioventricular) canal, one of these with pulmonary atresia. If you look at what has been published, these are patients who do not generally undergo complete repair as neonates. While there may be surgeons that would undertake that, I think that this is a reasonable alternative in such patients.
In terms of technical points, the distal end of the shunt is fairly straightforward: end-to-side anastomosis of the shunt to an opening in the central pulmonary arteries. For the proximal end, it is good to place it in a thin part of the RV (right ventricular) outflow tract. In a hypoplast, of course, you can figure that out by looking down through the pulmonary valve while you have the pulmonary artery open. In the current patients I didn't have a pulmonary artery to open, so it involved a little bit of guesswork. I just make a vertical incision with an 11-blade knife. I do make a point of looking inside to make sure that there aren't any muscle bundles crossing the area where I have made the incision, and if there are, I divide those. I don't resect muscle from the RV outflow tract, but when I make the subsequent anastomosis to the bevelled proximal end of the shunt I make a point of catching epicardium and endocardium in the suture line, which at least theoretically ought to compress the myocardium in the suture line and may prevent muscular ingrowth under the proximal end of the shunt.
In terms of the shunt clip, we have only used it in one hypoplast. This was a small baby who had pulmonary overcirculation at the end of an RV to PA shunt Norwood, or Sano operation, with a good result. He has now gone on to a successful second stage after having dilation in the cath lab.
DR JAMES S. TWEDDELL (Milwaukee, WI): This is an interesting concept and well presented. I have one comment and one question.
You addressed management of the oversized RV to PA conduit and I wanted to share with you our approach to that problem. The technique we have used is to take an additional segment of Gore-Tex graft and wrap it around that RV to PA conduit and fix it at several points with four or five stitches. The result is a longer segment of gentle narrowing. Like the technique of partial clipping this too is amenable to cath intervention. I just offer that up as an alternative to your clip technique.
Dr Ungerleider brought up the question of branch pulmonary artery rehabilitation. Narrowing of the proximal left pulmonary artery at the site of duct insertion is not rare in patients with pulmonary stenosis or atresia. This is a difficult problem in children palliated with a conventional systemic to pulmonary artery shunt. Did you identify any patients in your study group that had pulmonary artery coarctation at the site of duct insertion and were your interventionalists able to deal with that using this technique? You strategy is an interesting concept. There are certainly lesions that you might not want to tackle in the newborn period; furthermore, it may be worthwhile delaying placement of the initial right ventricular outflow tract valved conduit until the child is able to accommodate a larger initial conduit. Your data would suggest initial palliation with an RV to PA conduit is worth considering for some of these patients. Thank you.
DR BRADLEY: Thanks, Jim. Two of the patients had obvious left pulmonary artery problems at the time of the initial shunt placement; in one the PAs were discontinuous, and in those we fixed it at the time of the shunt placement. There was one patient, who during his follow-up developed proximal LPA stenosis, and this was successfully balloon dilated through the shunt in the cath lab. At the time of complete repair, we were placing a conduit that was quite a bit larger than the shunt, so any narrowing of the proximal branch pulmonary arteries was simply opened up across and the conduit placed into this opening.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. L. Owens and N. Musa Nutrition Support After Neonatal Cardiac Surgery Nutr Clin Pract, April 1, 2009; 24(2): 242 - 249. [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 |
