| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Division of Cardiovascular Surgery, Mayo Clinic, Rochester, Minnesota
b Division of Pediatric Cardiology, Mayo Clinic, Rochester, Minnesota
Accepted for publication April 14, 2008.
* Address correspondence to Dr Locker, Mayo Clinic, 200 First St SW, Rochester, MN 55905 (Email: lekerlocker.chaim{at}mayo.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: Thirty-two patients underwent EPAB; 20 patients had simultaneous creation of an APW. Median patient age was 40 days (range, 2 to 3,210); median weight was 3.5 kg (range, 2.4 to 23 kg). Endoluminal pulmonary artery banding fenestrations of 2 to 8 mm were centrally placed in a Dacron patch that was attached circumferentially and intraluminally in the main pulmonary artery. Fenestrations were sized by presence of APW and patient weight. Thirty-one of 32 patients underwent associated cardiac procedures. The mean follow-up period was 2.6 years (range, 0 to 15.5).
Results: Overall early mortality was 31% (10 of 32); 8% in EPAB alone (1 of 12) and 45% for EPAB+APW (9 of 20). Of the early deaths, 7 of 10 had severe, preoperative ventricular dysfunction. There was 1 early EPAB-related complication requiring band revision for relief of partial obstruction of the APW. At hospital dismissal, the mean pressure gradient after EPAB was 55.1 ± 8.4 mm Hg as assessed by echocardiography. No patient experienced distal pulmonary hypertension, distortion, or band occlusion. There were 6 late deaths. At late follow-up, 5 patients underwent band revision, and complete repair was accomplished in 10 patients.
Conclusions: Endoluminal pulmonary artery banding provided a consistently effective and durable reduction in pulmonary arterial blood flow with no pulmonary artery distortion. Early mortality was low for EPAB alone. Endoluminal pulmonary artery banding alone is preferred when controlled pulmonary blood flow and cardiopulmonary bypass are required to address intracardiac abnormalities. The role of EPAB with APW needs to be defined.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
Anomalies suitable for PAB in the modern era include multiple ventricular septal defects (Swiss cheese septum), anomalies destined to univentricular type repair (Fontan procedure), and patients in need of preparation of the subpulmonary ventricle for eventual arterial correction of transposition of the great arteries. Traditionally, the band is placed around the main pulmonary artery, and tightened as needed according to measurements of the pulmonary artery pressure distal and proximal to the band. The objective is to achieve a distal pulmonary artery pressure of one third to one half of the systemic blood pressure [2]. Extraluminal PAB can be associated with significant difficulties in trying to achieve the necessary degree of obstruction to pulmonary arterial blood flow. It may at times cause severe distortion of the pulmonary arterial confluence or the pulmonary valve [3]. We believe that placement of an endoluminal "band" may obviate some of these difficulties.
Furthermore, endoluminal pulmonary artery banding (EPAB) can be used with simultaneous creation of an aortopulmonary communication for the early palliation of complex anomalies with univentricular circulation. In this case, the band functions as a resistor to continuous pulmonary blood flow in effect replacing the function of a systemic-to-pulmonary artery shunt. This variation as an alternative to the Norwood approach is particularly attractive for patients with two normal-sized ventricles, with complex stenosis or atresia of the left ventricular outflow tract, for whom there is a realistic future expectation of a two-ventricle repair. The purpose of this study was to examine the results of this technique in our practice.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
|
|
Early mortality was defined as death occurring within 30 days of operation or at any time during the index hospitalization.
Cardiac Diagnoses
Patients selected for this procedure suffered from complex cardiac anomalies, and all patients had unrestricted pulmonary blood flow at birth. Patients selected for this procedure included anomalies not suitable for primary repair in infancy, anomalies being prepared for eventual Fontan procedure, and anomalies with two good-sized ventricles not suitable for primary repair because of complex left ventricular outflow obstruction. The predominant morphologic diagnoses are shown in Table 3. Other associated cardiac anomalies included patent ductus arteriosus in 19 patients, atrial septal defect in 10 patients, and bilateral superior vena cava in 10 patients. All patients had preoperative echocardiograms.
|
|
|
Pulmonary artery debanding was conducted during subsequent operations. The main pulmonary artery was divided at the EPAB site when a bidirectional cavopulmonary anastomosis was created, the distal main pulmonary artery was closed primarily, and the proximal MPA and pulmonary valve leaflets were closed separately. When a complete repair was performed, the EPAB was excised after opening the previous transverse incision, and the MPA was closed using a longitudinal oval-shaped patch.
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
The early mortality after EPAB and associated procedures was 31% (n = 10 patients); 7 patients were less than 10 days old. Nine of 20 patients (45%) in group 1 (EPAB with APW) died early, and 1 of 12 patients (8%) in group 2 (EPAB without APW) died. Seven of the 10 patients with early deaths went to operation with severe ventricular dysfunction, 1 of whom was in cardiogenic shock. Specifically, 3 of the early-death patients had moderate to normal-sized left ventricles with severe preoperative dysfunction, and 3 patients had severe right ventricular (systemic ventricle) enlargement and dysfunction with tricuspid regurgitation, with 1 requiring tricuspid annuloplasty. One patient was in cardiogenic shock preoperatively despite optimizing medical therapy, and operation was performed as a salvage procedure. Extracorporeal membrane oxygenation (ECMO) support was utilized in 2 of the 10 patients because of inability to separate from cardiopulmonary bypass. Morphologic diagnoses in the 10 patients who died early after operation are shown in Table 5. Eight patients died within 72 hours postoperatively, and the other 2 died on postoperative days 16 and 27. All 10 patients died of low cardiac output and multiple organ failure.
|
In all surviving patients, predismissal echocardiography demonstrated no evidence of pulmonary hypertension or distortion of the central pulmonary arteries. The postoperative echocardiographic estimation of the EPAB gradient was 55 ± 8 mm Hg. The mean systemic systolic blood pressure at the time of the echocardiogram was 91 ± 12 mm Hg. For surviving patients, the band resulted in significant improvement of the patient's clinical status, with no evidence of distal pulmonary hypertension nor distortion of the branch pulmonary arteries.
Twenty-two patients were followed for up to 15.5 years (mean, 2.6). There was no instance of occlusion of the endoluminal device. There were 6 late deaths. Three were related to sepsis, and most probably reflected the nature of their overall critical preexisting illness that led to multiple organ failure. A patient with hypoplastic left heart syndrome (HLHS) and double-outlet right ventricle died 6 months after EPAB and atrial septectomy owing to sepsis from respiratory complications. Another late death from pulmonary sepsis occurred 5 months after EPAB and APW, in a patient with HLHS and interrupted aortic arch. A patient with HLHS and double-outlet right ventricle, who underwent EPAB with APW, died of intracranial bleeding 6 months later, 3 weeks after undergoing a bidirectional cavopulmonary anastomosis. Another death occurred in a patient with HLHS, complete atrioventricular canal, APW, and interrupted aortic arch, who underwent EPAB and interrupted aortic arch repair. Extracorporeal membrane oxygenation support was established because of failure to wean from CPB; ECMO was discontinued successfully after revising the EPAB; the patient died of sepsis 5 months later. A patient with tricuspid atresia, transposition of the great arteries, and ventricular septal defect died 3 months after the initial procedure during reoperation for reduction of the size of the EPAB, mitral annuloplasty, and bidirectional cavopulmonary anastomosis. A patient with biliary atresia, HLHS, partial atrioventricular canal, and double-outlet right ventricle died 5 months after undergoing EPAB with APW; death was due to hepatic failure secondary to biliary atresia.
Ten of the 16 survivors underwent subsequent debanding and biventricular repair or Fontan procedure. Six of these patients underwent EPAB without APW, whereas 4 patients had both an EPAB and construction of APW. The median interval to debanding and complete repair was 2.4 years (range, 0.8 to 9) for patients without APW, and 1.7 years (range, 0.15 to 1.8) for patients with APW. The other patients are still palliated and awaiting their next operation. The last surgical procedures occurring after EPAB are outlined in Table 3.
Definitive repair (biventricular repair or Fontan) was achieved at the first operation after EPAB in 5 patients. Two underwent Fontan procedures, and 1 each a Kawashima repair, truncus arteriosus plus complete atrioventricular canal repair, and closure of ventricular septal defect.
The first reoperation after EPAB was palliative for 7 patients. Six patients underwent bidirectional cavopulmonary shunt, and 1 had construction of a modified Blalock-Taussig shunt. Other concomitant procedures were EPAB modification in 4 to adapt to somatic growth of the patient, aortic coarctation repair in 2, and mitral valve annuloplasty, aortopulmonary window enlargement, new aortopulmonary window, and ligation of a persistent left superior vena cava in 1 patient each.
During the second reoperation, complete repair was achieved in an additional 5 patients. Two patients underwent a modified Fontan procedure, 1 patient underwent a Konno procedure, 1 patient underwent closure of ventricular septal defect with reconstruction of the right ventricular outflow tract, and 1 patient had left ventricular outflow enlargement with closure of ventricular septal defect and APW.
|
Comment |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
Complications of conventional (external) PAB include erosion and migration of the pulmonary band resulting in inadequate stenosis of the pulmonary artery with pulmonary hypertension, distal migration resulting in distortion and stenosis of the branch pulmonary arteries, or impingement on the pulmonary valve apparatus [10].
Since the first report of the technique of PAB by Muller and Damman [1], many clinical and experimental techniques for extraluminal PAB have been introduced. They include banding the main pulmonary artery with percutaneously adjustable prosthetic devices [11, 12], the use of prosthetic materials that can undergo balloon dilatation postoperatively [13, 14], the use of absorbable materials for the PAB [15, 16], and adjustable banding devices that can be adjusted remotely [17]. The endoluminal PAB technique was first described by Doty and coworkers [18] in a case report as a palliative measure in a patient with HLHS [18]. This report was followed by isolated case reports of various combinations of EPAB [19]. Piluiko and colleagues [20] reported a retrospective series of 18 patients who underwent EPAB with a pericardial patch, over a 10-year study period. Their hospital mortality after the EPAB operation was 22%, and for the debanding procedure, it was 15.4%. Thirteen (93%) of their 14 survivors underwent pulmonary artery debanding at a median interval of 5.9 months (range, 3.7 to 12.4). The authors reported that the EPAB resulted in significant and consistent reduction in the systolic pulmonary artery pressure. Their patients had no band-related anatomic complications but 2 patients required percutaneous dilatation before debanding as palliation for systemic arterial desaturation.
In our practice, the endoluminal device was used as a substitute for the traditional, extraluminal PAB in patients with congenital malformations who required additional intracardiac palliative procedures requiring CPB. All of these patients had anomalies not suitable for primary definitive repair. Satisfactory placement and sizing of an extraluminal PAB after a palliative procedure requiring CPB is difficult. In the absence of left ventricular outflow tract obstruction, the technique of EPAB is simple and safe (early mortality < 10%), and resulted in a predictable reduction in pulmonary artery blood flow and pressure, with no late band-related complications. The length of time to place the EPAB while on bypass was 10 to 15 minutes for all cases. Consequently, we believe that EPAB is the preferred method for controlling unrestricted pulmonary blood flow when the LVOT is adequate and other intracardiac procedures are required.
In addition, the device was also used in combination with a surgically constructed APW as a preferred alternative to the Norwood operation in patients with complex left ventricular outflow obstruction, in whom the size of the left ventricle was below normal, but had at least some potential for subsequent biventricular repair. It was also used in patients with functional single-ventricle circulations who had the substrate of subaortic stenosis, as an alternative to the Damus-Kaye-Stansel connection with isolation of the pulmonary confluence and placement of a systemic-pulmonary artery shunt. The EPAB obviated the need for the conventional systemic–pulmonary artery shunt with its propensity for thrombosis and pulmonary arterial distortion and stenosis. There were no isolated Damus-Kaye-Stansel procedures performed during this review as the EPAB-APW was our preferred procedure for this constellation of anatomical defects. However, the early mortality rate in this group was notably high. The procedure was first performed (1990) when the mortality rate was still high for the Norwood procedure, which would have been the alternative procedure most likely to have been performed. Early in this experience, we had particular enthusiasm for the EPAB-APW procedure instead of the Norwood procedure when there appeared to be potential for an eventual biventricular repair since many of these patients had nondiminutive left ventricles. In fact, 6 of the 9 patients who died in this group could have been treated with a Norwood-type operation. However, the risk of the Norwood operation would have also been increased because many of the patients who died early in this series had significant preoperative complicating features. They included 3 patients with moderate to normal-sized and severely dysfunctional left ventricles and 3 patients with severe right ventricular (systemic ventricle) enlargement and dysfunction with concomitant tricuspid regurgitation. In addition, 1 patient was in cardiogenic shock despite attempts at stabilization with medical therapy, and operation was performed as a salvage procedure. Even with the improvements in perioperative intensive care unit care and refinements in intraoperative management, these patients who died early would have had a high mortality rate with any conventional operation, even in the current era. With further refinements in ECMO support programs and with more liberal application of ECMO in the postcardiotomy setting when there is difficulty separating from cardiopulmonary bypass, there likely would have been further improvement in the overall mortality of this group of patients.
From this small cohort of patients it is difficult to draw definitive conclusions about which patients should or should not be treated with EPAB and APW. Our sense is that the patients with a HLHS complex diagnosis (especially those with nondiminutive, dysfunctional left ventricles) did poorly compared with other diagnoses, but the severely compromised preoperative conditions and ventricular dysfunction of many of those patients make analysis difficult. While transplantation is also an alternative for this constellation of anatomic abnormalities, it is often not practical, given limited donor availability. The improved early results with the Norwood operation in the current era may make this a better initial option for palliation in patients with inadequate left ventricular outflow tracts and a need for controlled pulmonary blood flow. Patients with functional single-ventricle anatomy and the potential for developing subaortic stenosis who do not require arch reconstruction may be the best subgroup for EPAB and APW palliation.
Late follow-up was obtained in 14 patients. All patients reported their health status as fair to excellent, with 80% reporting good to excellent. Two patients reported increase in cyanosis resulting from somatic patient growth. In 10 of the surviving patients, complete surgical repair was achieved during the study period.
In our experience, the EPAB as a restrictor for pulmonary arterial blood flow without a surgically created APW is an effective technique resulting in minimal or no distortion of the distal and proximal architecture of the pulmonary arteries and pulmonary valve (Fig 2). The EPAB produced a predictable and reproducible reduction in pulmonary blood flow. Endoluminal pulmonary artery banding provides an effective palliative procedure in preparation for definitive univentricular or biventricular repair of selected cardiac anomalies. Given the improved results of the Norwood operation, the role of APW with concomitant EPAB is limited in the current era. In the absence of significant preoperative risk factors, its application remains to be defined.
|
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
As you showed us, there are basically two groups, those who had band alone and those who had band plus AP window. I think in your band only group, at least I would agree, you can do this procedure, and you can do it safely. You don't have the distortion, the things you are worried about, by extraluminal banding. My interest more is in that second group, AP window and pulmonary artery banding, and what I would consider an exceedingly high mortality of 45% in that group. So my questions are related to that group of patients.
Were there any data in that group about the effectiveness of the band? You showed us in the patients who go home, they have a band gradient of 55 mm Hg. In those patients who died, do you have any data that that band was as effective? I mean, I assume, or I would guess, that it is difficult to titrate the band in a small patient. And in fact, 7 of your patients who died were less than 10 days old. These may be the sickest patients, these may be the patients with hypoplastic left heart syndrome, but are there any data that we know that that band was effective? In Doty's first report, exactly what you are doing here, when they catheterized their patients, the Qp/Qs was 3 to 1 using a similar size fenestration that you do. So if you have a sick ventricle and you leave the patient with a 3-to-1 shunt, he is not going to do well. So the first question is, do you have any data in that group that your band was effective in sufficiently limiting pulmonary blood flow?
And then secondly, just to put these results in perspective, at the very beginning of this meeting at the postgraduate course, we talked about hypoplastic left heart syndrome, and in many centers now, the results for mortality after Norwood palliation, ought to be probably 10% or so; this is what it seems like big centers are getting toward. So many of your patients could have been treated with a Norwood with a mortality of 10%, maybe, rather than 45%. There is a recent paper from the United Kingdom this year where they looked at palliating hypoplastic left heart syndrome with an AP window and conventional pulmonary artery band and found just what you did. Their mortality was 54%, plus 2 more patients they converted back to a Norwood type palliation. Their feeling being that there is a systemic "steal" or low diastolic pressure as the cause of the high mortality. So with your data, with their data, and with the current data related to Norwood survival, do you feel that this is a useful technique in this subgroup of patients who could be palliated with a Norwood type palliation? Thank you very much. Nice presentation.
DR DEARANI: Thank you, Dr Heinle. Your first question has to do with sizing of the fenestration and Qp/Qs. The fenestrations were undersized in patients who also had an aortopulmonary window in an effort to avoid pulmonary over circulation. In these patients, the band functions like a shunt with pulmonary blood flow occurring during systole and diastole. In these patients, the oxygen saturations were in the 75% to 80% range in the early postoperative period, so overcirculation was unlikely. Fenestrations were slightly larger in those patients who had a band without a concomitant aortopulmonary window.
With regard to your second question, many of the patients with aortopulmonary window were done in the early 1990s. At that time, this procedure was done as an alternative to the Norwood procedure because the results with the Norwood at that time were not very good, and, in addition, they were done in patients where it was felt that there was a potential for a subsequent two-ventricle repair. Most of us have had the experience that two-ventricle Fontans don't do particularly well, especially if one ventricle is functioning poorly. Since our review of the results of the procedure presented here, we have moved back to doing the Norwood procedure, although the analysis of late outcome for Norwood staged palliation in the setting of two ventricles will be necessary to determine the optimal procedure. It is not clear to us whether the poor results obtained in this series, is procedure related or patient related.
In our current practice, we consider heart transplantation for those patients with two ventricles with one that is functioning poorly if the family will give consent to it. If conventional operation is performed, we proceed with Norwood palliation and utilize ECMO support postoperatively.
DR ROSS M. UNGERLEIDER (Portland, OR): This is an interesting concept. It takes a lot of courage to present it to this critical audience. Has this ever been described before or, as far as you know, is this the first time that somebody has done this? And my questions to you are three more.
Be more specific, if you would, for us about how you size the fenestration in terms of millimeters in this Gore-Tex patch? Number two, it has been my experience banding patients who sometimes we have to adjust the band. Is there any way with your technique that you can adjust the size of this fenestration? Has that been an issue for you in any patients? And third, have you talked to some of the interventionalists about putting in an intraluminal device without bypass in the patients who don't need left ventricular outflow procedures as well? Then it might really be attractive. This sounds like something that would really fit into the hands of an interventionalist. Thank you.
DR DEARANI: The first question relates to sizing of the endoluminal pulmonary artery band. For patients who had pulmonary artery banding for the sole purpose of banding, the size of the fenestration was half the size of the aortic annulus. In patients who had it in conjunction with aortopulmonary window so it was functioning as a shunt as well, the size was slightly smaller and was 2 and 3 mm for newborns, 3 to 4 mm for infants. In a newborn, we typically used a 2.7 mm punch. Concomitant intracardiac procedures requiring cardiopulmonary bypass were also performed. To the best of my knowledge, I think this concept of banding with concomitant aortopulmonary window or a Damus-Kaye-Stansel type of communication was Dr Puga's idea.
DR JEFFREY JACOBS (St. Petersburg, FL): Joe, I don't rise for questions. I just have two quick comments. First, I congratulate you for presenting an innovative approach and making a very honest presentation of the results with that approach. That is commendable. And second, there has been a lot of discussion over the last few days about centers with 10% mortality, discharge mortality, after a stage 1 Norwood. I think it is important to realize in the STS database, which is the largest multi-institutional collection of patients undergoing stage 1 palliation, over the last several years the discharge mortality has been more along the lines of 20% to 30% to 35%, and in the last few reports, 20% to 30%, and that includes a few centers that are reporting 10% discharge mortalities. So across the country, certainly the average outcome is not a 10% discharge mortality for a stage I Norwood. Thanks.
DR CONSTANTINE MAVROUDIS (Chicago, IL): This was an excellent presentation. While I do not have a question, I do have a comment concerning an historical account of intraluminal pulmonary artery banding in a patient with transposition of the great arteries and intact ventricular septum who presented too late for arterial switch. While I was visiting Bergamo, Italy, in about 1991, Lucio Parenzan and his colleagues performed transcatheter intraluminal balloons for pulmonary artery banding for about a week, which was followed by arterial switch. I'm not sure of the result and I'm not sure that this was published. I rise to comment that the idea of an intraluminal pulmonary artery band has been attempted in the past, albeit temporarily for left ventricular training. Congratulations on your paper.
DR DEARANI: Thank you. There was one question that I forgot to answer for Dr Ungerleider, that related to the need for cardiopulmonary bypass. It is important to emphasize that all of these children required an open operation to address other intracardiac pathology at the time of banding. Consequently, the alternative of a percutaneous approach would not have been applicable for this particular group of patients. In the absence of intracardiac problems that would require an open operation, the percutaneous approach or simple extraluminal pulmonary artery banding would have been an option to consider. Thank you.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. Eme, J. Gwalthney, J. M. Blank, T. Owerkowicz, G. Barron, and J. W. Hicks Surgical removal of right-to-left cardiac shunt in the American alligator (Alligator mississippiensis) causes ventricular enlargement but does not alter apnoea or metabolism during diving J. Exp. Biol., November 1, 2009; 212(21): 3553 - 3563. [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 |
