| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J Thorac Cardiovasc Surg 1997;114:718-726
© 1997 Mosby, Inc.
SURGERY FOR CONGENITAL HEART DISEASE |
Received for publication May 7, 1997 revisions requested June 12, 1997; revisions received July 11, 1997 accepted for publication July 17, 1997. Address for reprints: V. Mohan Reddy, MD, 505 Parnassus Ave., M589, San Francisco, CA 94143-0118.
Abstract
Background: A modified Damus-Kaye-Stansel procedure is one of several options for palliation of single ventricle with subaortic obstruction, but results in neonates have been disappointing. In the presence of arch obstruction, this procedure is typically performed with circulatory arrest, which may contribute to neurologic insult. Methods: Since 1990, a modified Damus-Kaye-Stansel procedure has been performed in 14 neonates and seven infants with single ventricle and subaortic stenosis, including 15 with arch obstruction. Diagnoses were double-inlet left ventricle (n = 12), tricuspid atresia (n = 2), and other forms of hypoplastic ventricle with subaortic obstruction (n = 7). Three patients underwent concurrent bidirectional Glenn shunt. In the most recent seven patients with arch obstruction, arch repair was achieved with an end-to-side anastomosis of the descending aorta to the ascending aorta with continuous upper body perfusion. Results: One early death occurred among the 14 neonates (7%) and three among the infants, for an early mortality of 19%. At a median follow-up of 33 months, there were no late deaths or neurologic complications. Nine patients underwent subsequent bidirectional Glenn anastomosis, including three who had Fontan completion and one who later underwent conversion to a partial biventricular repair. One patient required a transplant for cardiomyopathy of unknown etiology. The remaining 12 patients are considered good candidates for Fontan completion. No patient has recurrent arch obstruction. Four patients have mild (n = 1) or trivial (n = 3) semilunar valvular regurgitation. Conclusion: The modified Damus-Kaye-Stansel procedure is an effective primary palliation for single ventricle and subaortic stenosis, with or without arch obstruction. Results are especially encouraging in neonates. Arch repair can be achieved without circulatory arrest to the brain .
Anastomosis of the proximal end of the transected main pulmonary artery to the ascending aorta, described independently by Damus,
1 Kaye, 2 and Stansel, 3 may be used in the palliation or repair of a variety of lesions. Norwood's operation for stage I palliation of hypoplastic left heart syndrome 4 is the most powerful example of how this concept has affected the management of complex congenital heart disease. In addition, modifications of the Damus-Kaye-Stansel (DKS) procedure are also useful in univentricular transposition complexes such as {S,L,L} double-inlet left ventricle and {S,D,D} tricuspid atresia, in which the aorta arises from a rudimentary outflow chamber via a bulboventricular foramen (or ventricular septal defect) that is frequently restrictive. 5-11 In such lesions, unlike with hypoplastic left heart syndrome, neonatal palliation by main pulmonary artery–ascending aorta anastomosis is not always necessary, because critical subaortic stenosis is not always present during the first few months of life and other methods of primary palliation are available. 12-15 Because early outcomes of the DKS procedure in neonates and infants have generally been poor, 5,6,11 options such as pulmonary artery banding with or without arch repair are often used for first-stage palliation. 12,15-18 Evidence exists that pulmonary artery banding can precipitate earlier development of significant subaortic obstruction caused by both muscular hypertrophy 15 and volume unloading 19 and that patients with arch obstruction are at particular risk. 16,20 Subsequent palliation may be compromised by the banding procedure when subaortic obstruction develops, making enlargement of the bulboventricular foramen the only option to relieve the systemic outflow obstruction. Bulboventricular foramen enlargement may have a negative impact on rhythm and ventricular function. Such developments may ultimately render the patients suboptimal candidates for the Fontan operation. 20-23 Thus main pulmonary artery–ascending aorta anastomosis in the neonatal period may be the most beneficial primary palliation in many neonates with forms of functional single ventricle and actual or potential systemic outflow obstruction other than hypoplastic left heart syndrome, so long as early mortality can be reduced and significant morbidity, especially neurologic insult and semilunar valvular dysfunction, can be avoided. In this report, we describe our experience with modifications of the DKS procedure, with an emphasis on techniques that avoid great vessel distortion and subsequent semilunar valve dysfunction and that also avoid deep hypothermic circulatory arrest and the associated neurologic insult.
Patients and methods
Patients.
Between May 1990 and April 1997, anastomosis of the main pulmonary artery to the proximal ascending aorta was performed in 21 neonates and infants with forms of functional single ventricle other than hypoplastic left heart syndrome. Patients undergoing this procedure beyond 1 year of age were not included in this study, nor were patients in whom the DKS operation was performed for biventricular repair, as in Taussig-Bing double-outlet right ventricle. Fourteen of the 21 patients were 1 month of age or younger, 15 had significant arch obstruction, and 17 had undergone no previous palliative procedures. Patients fit into three diagnostic categories: {S,L,L} double-inlet left ventricle (n = 12; Table I), {S,D,D} tricuspid atresia (n = 2; Table II), and other forms of functional single ventricle (n = 7; Table II). All patients had actual or potential subaortic obstruction according to either a bulboventricular foramen index of 2.0 cm2/m2 or less, as proposed by Matitiau and associates, 20 a bulboventricular foramen/aortic valve diameter ratio of 1 or less, or a systemic outflow gradient at the subvalvular level. Among the 12 patients with double-inlet left ventricle were 11 neonates, and no patient had undergone a prior palliative operation. Arch obstruction was present in 10 of these patients. The two patients with tricuspid atresia (3 and 7 months of age) were without arch obstruction, and one had undergone prior pulmonary artery banding and atrial septectomy. Only three of the seven patients with other forms of single ventricle were neonates, and three had undergone prior operations.
|
|
24 In patients with a markedly hypoplastic ascending aorta, the transverse leg of the incision was typically very short. The main pulmonary artery was transected above the sinotubular junction, and the portion of the proximal stump abutting the ascending aorta was anastomosed to the transverse component of the aortic incision. The flap of aortic tissue subtended by the aortotomy was retracted posteriorly, and the pulmonary artery–aorta anastomosis was continued with the lower edge of the aortic flap sewn to the adjacent edge of the pulmonary artery stump. A patch of either allograft or pericardium was incorporated as an anterior hood to complete the anastomosis (Fig. 1). In the other nine patients, both the ascending aorta and main pulmonary artery were transected above the sinotubular junction, and the proximal stumps were anastomosed side to side, incorporating approximately 30% to 35% of the circumference. The distal portion of the ascending aorta was opened longitudinally and anastomosed end to end to the proximal great vessel confluence, with augmentation of the anastomosis and ascending aorta by means of a patch of allograft tissue. Five of the patients who underwent this procedure did not require arch repair, and the allograft patch was used only to augment the ascending aorta. The other four were patients with arch obstruction (including two neonates with interrupted arch) in whom the longitudinal incision in the ascending aorta was extended across the arch and isthmus and augmented with a patch of allograft tissue as in the stage I Norwood procedure. In patients with L-transposed great arteries who underwent this procedure, the augmented ascending aorta was transposed rightward and sewn end to end to the previously anastomosed proximal main pulmonary artery and proximal ascending aorta.
In the most recent seven patients with arch obstruction, arch repair was achieved by performing an end-to-side anastomosis of the descending aorta (distal to the ductus) to the ascending aorta, after resection of all ductal tissue (Fig. 1). This technique was similar to that which we use for standard repair of interrupted arch from a midline approach. A crossclamp was placed immediately proximal to the aortic inflow cannula, at the base of the innominate artery. A longitudinal incision was made along the lateral aspect of the ascending aorta just proximal to the clamp, and the descending aorta was anastomosed end to side to the ascending aortotomy opposite the DKS anastomosis. The levels of the DKS anastomosis and the descending aorta–ascending aorta anastomoses were overlapping, creating an almost direct path from the pulmonary trunk into the descending aorta. Upper body perfusion was maintained by cannulating at the base of the innominate artery, distal to the clamp (Fig. 1), snaring the proximal left common carotid and subclavian arteries. Perfusion was maintained at a rate of 30 ml/kg per minute with an alpha-stat approach and no change in the priming solution. In one of these patients (patient 21) with a very unusual anatomy, including mesocardia, muscular subaortic obstruction, coarctation of a right-sided arch along with a left-sided descending aorta, and a dominant coronary artery arising significantly above the sinuses of Valsalva, it was necessary to perform the DKS anastomosis more distally than usual to avoid the anomalous coronary artery. The descending aorta was anastomosed end-to-side to the superior aspect of the flap of autologous aorta used for the DKS anastomosis, and the three-way anastomosis was completed with a hood of allograft tissue.
Additional procedures.
All 17 patients in whom the DKS procedure was performed as a primary palliation had a 3 or 3.5 mm shunt placed from the innominate artery to the branch pulmonary artery to supply the pulmonary circulation. In three patients in whom a DKS procedure was performed beyond the neonatal period (patients 2, 15, and 19), a bidirectional Glenn anastomosis was placed during the same operation. In one neonate with double-inlet left ventricle (patient 9), a severely regurgitant left atrioventricular valve was oversewn. In all three patients with previously placed pulmonary artery bands and the one patient who had initially undergone biventricular repair of an unbalanced atrioventricular septal defect. The bands were taken down and the atrioventricular septal defect repair reversed.
Cardiopulmonary bypass and cardioplegic arrest were used in all patients, with a median perfusion time of 152 minutes (range 86 to 200 minutes) and a median crossclamp time of 47 minutes (range 30 to 98 minutes). Total circulatory arrest ranging from 45 to 60 minutes was used in the five patients who underwent a Norwood type of arch repair. In the seven patients who underwent arch repair with end-to-side anastomosis of the descending aorta to the ascending aorta, circulatory arrest to the brain was not used. Rather, the aorta was clamped just proximal to the innominate artery, allowing perfusion of the head vessels by inserting the inflow cannula at the base of the innominate artery, as described earlier.
Data collection.
Perioperative data were collected on retrospective review of patient records. Follow-up was conducted by direct contact with referring physicians and was current and complete in all patients.
Results
Early outcomes.
Among the 14 neonates who had first-stage palliation with this procedure, there was one early death (7%). This was a patient with double-inlet left ventricle who underwent surgery early in our experience, with the use of circulatory arrest. He had a cardiac arrest on the day of the operation, and autopsy showed evidence of acute myocardial infarction and neurologic damage, including recent and remote cerebral necrosis. Three additional deaths occurred in patients who underwent a modified DKS procedure beyond the neonatal period for palliation of forms of univentricular heart other than {S,L,L} double-inlet left ventricle (
Table II). One of these patients had previously undergone biventricular repair for unbalanced atrioventricular septal defect with subsequent takedown and a modified DKS procedure as a salvage procedure when it became evident that the left ventricle was not adequate. Another of these deaths occurred in a patient with asplenia syndrome and subvalvular obstruction of both outflow tracts who had an unexplained cardiopulmonary arrest 2 days after the operation and could not be resuscitated even with a ventricular assist device. The third death occurred in a patient who underwent primary palliation at 94 days of age for tricuspid atresia. This patient had a coagulopathic reaction and volatile cardiopulmonary status after the operation, which was consistent with a transfusion reaction. Cardiac arrest occurred the night after the operation, and she could not be resuscitated. There were no known neurologic events before hospital discharge. Electroencephalography was not performed, but there was no clinical evidence of seizure activity in any patient.
Reoperation for bleeding on the day of the operation was performed in one neonate. Otherwise, no patients required reoperation in the early postoperative period. Mechanical ventilatory support was continued for no longer than 7 days in any patient, and the median postoperative hospital stay was 17 days (range 4 to 42 days).
Follow-up.
No late deaths occurred during a follow-up period ranging from 1 month to 6.8 years (median 33 months). Nine patients underwent second-stage palliation with a bidirectional Glenn anastomosis between 4 and 11 months (median 5.9 months) after construction of the modified DKS anastomosis. At this time, eight patients were without obstruction to systemic outflow at any level. One patient who had undergone a Norwood type of arch reconstruction had a mild gradient at the level of the distal arch, which was augmented with a patch of allograft tissue. None of these patients required additional operations between the modified DKS and bidirectional Glenn procedures. In all, three patients have had the Fontan circulation completed at 20, 28, and 46 months after the bidirectional Glenn anastomosis. Another patient (patient 16) underwent partial biventricular repair 33 months after modified DKS anastomosis and concurrent bidirectional Glenn anastomosis. The DKS anastomosis was taken down, the pulmonary outflow tract reconstructed, and complete intracardiac repair performed, with maintenance of the bidirectional Glenn shunt. Another patient (patient 17) required cardiac transplantation for progressive acquired cardiomyopathy of unknown origin 19 months after successful neonatal DKS palliation. All remaining patients (n = l2) are currently considered good candidates for the Fontan operation; three are awaiting intermediate palliation with a bidirectional Glenn anastomosis and the others are awaiting Fontan completion.
Of the 32 semilunar valves among the 16 surviving patients (excluding the patient who received a transplant), follow-up echocardiography showed no regurgitation in 28 valves, trivial regurgitation in three (two aortic and one pulmonary), and mild regurgitation in one. One of the patients with trivial aortic regurgitation was a 4.8-month-old patient with a 90 mm Hg subaortic gradient and trivial-mild aortic regurgitation before the operation. No valves have moderate or severe insufficiency. All patients have good ventricular function except for one, in whom mild-to-moderate ventricular dysfunction had developed before bidirectional Glenn shunting. This was improved after the bidirectional Glenn anastomosis, but mild depression remained at follow-up 2 months later. Three patients had mild insufficiency of the small right atrioventricular valve. None of the patients who underwent arch reconstruction with descending aorta–ascending aorta anastomosis has evidence of obstruction to the aortic arch, left pulmonary artery, or tracheobronchial tree.
|
Discussion
In treating the neonate who has functional single ventricle, optimization of the physiology for an ultimate Fontan procedure should be of high priority even in the earliest stages of management. Factors that have been established as contributing to poor outcome after the Fontan operation include ventricular hypertrophy or other ventricular dysfunction, elevated pulmonary artery pressure and pulmonary vascular resistance, and significant valvular dysfunction. 21-23
In neonates who will require palliation of functional single ventricle, therapy should be designed to minimize the development of these processes. In patients with actual or potential obstruction to systemic outflow in conjunction with unrestricted pulmonary blood flow, a modified DKS procedure (with or without arch repair) is an attractive option for primary palliation, inasmuch as it provides relief at all levels of systemic obstruction along with a controlled source of pulmonary blood flow. 5,16,13 Since this application of the ascending aorta–main pulmonary artery anastomosis was first reported by Yacoub and Radley-Smith, 25 however, the primary drawback has been high early mortality. 5,6,11 A secondary drawback has been the development of semilunar valvular dysfunction. 7,8,26,27 Another concern with the DKS procedure, especially in the presence of arch obstruction, is that it has generally been performed with deep hypothermic circulatory arrest, which is known to cause neurologic insult. 28,29
Another common approach to the single ventricle with subaortic obstruction is to band the pulmonary artery early in life along with arch repair if necessary. An attractive feature of this approach is that it avoids brain ischemia, cardiopulmonary bypass, and the associated morbidity. Although several recent studies have found early mortality to be low after pulmonary artery banding in neonates with this type of lesion, 17,18 other series have reported outcomes that were not substantially different from those achieved with ascending aorta–main pulmonary artery anastomosis. 14 Moreover, substantial evidence exists that obstruction to aortic outflow is likely to develop after pulmonary artery banding in this group of patients. Initially, it was proposed that progressive ventricular hypertrophy led to relative narrowing and consequent obstruction of the bulboventricular foramen. 15,16 More recent studies have demonstrated that ventricular geometry is altered early after volume reduction surgery, including pulmonary artery banding, with a resulting increase in wall thickness and bulboventricular foramen narrowing. 19 In the setting of prolonged pulmonary artery banding in a single ventricle pumping systemic and pulmonary output in parallel, both of these mechanisms are probably contributory when subaortic obstruction does develop.
The risk of developing progressive subaortic obstruction after primary pulmonary artery banding for {S,L,L} double-inlet left ventricle and {S,D,D} tricuspid atresia is not the only factor offsetting the generally low early mortality after this approach. Because outflow obstruction will inevitably develop in most patients, 15,17 some method of relief (either a modified DKS procedure or bulboventricular foramen enlargement) must be obtained either before second-stage palliation or simultaneous with bidirectional Glenn anastomosis or Fontan completion. In some centers' experience, this might necessitate an additional operation in infancy or early childhood. When subaortic obstruction is addressed after the Fontan operation, mortality can be very high. 30 Even though the operative risk of a modified DKS procedure has been lower among older patients than among neonates and infants in previous series, there is still a mortality of at least 10% in most reports. 9,10,12 In addition, pulmonary artery band complications, such as migration causing pulmonary artery or valve distortion, are not uncommon and may eliminate the DKS procedure as an option, thus leaving bulboventricular foramen enlargement as the only resort to relieve systemic outflow obstruction. A number of objections to the use of bulboventricular foramen enlargement in these patients have been expressed. Removing significant amounts of ventricular muscle, when ventricular function is so central to long-term outcome, must be viewed as an inferior option, in our opinion. Furthermore, despite better understanding of the conduction tissue in these patients with complex anomalies, surgically induced complete atrioventricular block in association with enlargement of the bulboventricular foramen remains a real possibility. It is our belief that staying out of the ventricle is almost always the best option in patients with single ventricle.
In the present report our focus is on two groups of patients: (1) those with double-inlet ventricle and a high likelihood of subsequent subaortic obstruction and (2) another group of patients who were generally beyond the neonatal period, had undergone prior palliation, and who often had unusual forms of complex single ventricle with subaortic stenosis already present. In several series of patients undergoing either main pulmonary artery–ascending aorta anastomosis 6 or pulmonary artery banding 16 as first-stage palliation for univentricular transposition complexes, arch obstruction has been associated with poor early outcome. Patients with single ventricle and arch obstruction are also more likely to have or develop subaortic obstruction. In our cohort of neonatal patients with double-inlet left ventricle, early and late outcomes have been excellent. There was one early death after a modified DKS procedure, but all other patients are alive and well, most having undergone second- or third-stage palliation.
In the older group of patients in this study, results have been less encouraging. This group was extremely heterogeneous in terms of morphologic characteristics, physiologic presentation, and clinical setting. The surgical options for these patients (Table II) were fewer, inasmuch as most had an actual subaortic gradient detected, had already undergone primary palliation, or had lesions that were not amenable to pulmonary artery banding. Thus this group of patients is not one in which palliation with another less extensive procedure would have reduced mortality. Of the three infants in this group who died, one underwent a secondary DKS procedure as a salvage attempt after failure of biventricular repair for unbalanced atrioventricular septal defect, another had asplenia syndrome with the very unusual finding of subaortic and subpulmonic obstruction, and the third died after a transfusion reaction.
Both modified DKS and pulmonary artery banding have clear indications for palliation of functional single ventricle hearts. Echocardiographic studies in patients with univentricular transposition complexes have identified several factors that may predict the development of subaortic obstruction. 20 However, even if such data are taken into account, there is a relatively large group of patients in whom the adequacy of the subaortic region for supporting unrestricted outflow to the systemic circulation is equivocal. The development of techniques as described in the present report, which avoid deep hypothermic circulatory arrest and can provide very low neonatal mortality and excellent long-term valvular function, may shift the balance in favor of modified DKS in many of the unclear cases. On the basis of our experience, we believe that a DKS procedure should be used whenever the possibility of subaortic obstruction is present, unless significant semilunar valvular regurgitation is present. In patients with arch obstruction and an ascending aorta smaller than about 4 mm in diameter, we would be more likely to repair the arch with circulatory arrest or a modified Norwood procedure than with the technique described in the present report. Pulmonary artery banding should be used only when there is no reasonable likelihood of bulboventricular foramen obstruction, which we determine on the basis of a bulboventricular foramen area larger than the aortic valve area, a bulboventricular foramen index greater than 2 cm2/m2, and no potential obstruction in the subaortic outlet chamber. Enlargement of the bulboventricular foramen should be avoided whenever possible.
Appendix: Discussion
Dr. Thomas L. Spray(Philadelphia, Pa.).
I congratulate the authors on an excellent series with superior results in complex arch repairs in infancy.
We and others have reported results with pulmonary banding and early takedown to a DKS and bidirectional Glenn or a hemi-Fontan procedure with results comparable with those presented in this series. However, in the neonate with significant subaortic obstruction associated with single ventricle, it seems appropriate to use arch reconstruction and aortopulmonary shunting as the primary operation to avoid the potential effects of pulmonary banding on ventricular hypertrophy and consequent progression of subaortic obstruction. As experience with neonatal arch reconstruction associating the main pulmonary artery and ascending aorta has increased, primary operative intervention can now be undertaken with satisfactory results. The technical modifications of the aortic root reconstruction with a variation of the DKS operation, as described by these authors, is certainly one of many potential approaches. Others include a Norwood-type arch reconstruction or direct pulmonary artery–arch anastomosis, as has been advocated by Dr. Mee and Dr. Brawn. The spectrum of anatomy in these infants with functional single ventricle and subaortic stenosis with varying degrees of aortic arch obstruction would suggest that an individualized application of alternative techniques may be the most appropriate approach.
The very low incidence of semilunar valve insufficiency in patients in this series would suggest that neonatal intervention and reconstruction may be associated with less distortion of the pulmonary or aortic valves than might be seen from pulmonary banding and might perhaps optimize late results.
In children with functional single ventricle and significant subaortic stenosis associated with arch hypoplasia, obstruction, or interruption in the neonatal period, we have elected to use a Norwood-type reconstruction of the entire aortic arch, as was described for a small number of the patients in the authors' series. In the 2-year period of 1995 and 1996, 26 consecutive patients underwent such reconstructions. In patients with transposition and tricuspid atresia, division of the proximal aorta and pulmonary artery was performed with arch augmentation and reconstruction proximally, as described by the authors. In situations in which arch interruption was present, extensive mobilization of the arch vessels with extension of incisions into the origin of the carotid and subclavian vessels to create the widest possible aortic arch anastomosis was performed with associated homograft arch augmentation. This was often performed with division of the great vessels, as in those patients with transposition. In infants in whom the proximal great vessels were suitably aligned, a Norwood-type direct augmentation of the arch with proximal anastomosis of the aorta to the pulmonary artery was performed. The overall mortality in this 26-patient group was 15% in the neonatal period.
The technical modifications suggested by the authors of this study are intriguing. In patients with a very small ascending aorta, as in those with arch obstruction, we have found it technically difficult to create a DKS connection proximal to the takeoff of the innominate artery and avoid tenting of the distal anastomosis from the pulsatility of cardiac ejection, since the distance is often short between the ascending aorta and the takeoff of the innominate vessel. The use of arch augmentation creates a widely unobstructed arch with less potential for distortion in our experience. The authors are to be commended for the excellent late results with their approach. In addition, we have been concerned about direct anastomosis of the descending aorta to the side of the ascending aorta proximal to the innominate artery, because even with extensive mobilization, compression of the left main-stem bronchus or left pulmonary artery could occur with implications for respiratory complications or growth limitation of the left pulmonary artery. Have the authors noted any compromise of left pulmonary artery size by arch reconstruction in their patients who are awaiting the completion Fontan procedure?
The use of continuous cardiopulmonary bypass through the base of the innominate artery certainly minimizes the need for circulatory arrest for arch reconstruction. Dr. Asou and associates from Kyushu University in Japan have described similar selective cerebral perfusion strategies for arch reconstructions in neonates. They have described an intriguing use of the proximal anastomosis of the aortopulmonary shunt created to the origin of the innominate artery for perfusion of the innominate vessel while arch reconstruction is being performed. Have the authors used this technique, which would seem to be technically advantageous and would avoid the problems of cannulating a very small innominate artery or ascending aorta? Potential narrowing of the vessel at decannulation could limit inflow into the aortopulmonary shunt with the authors' technique.
Dr. Shunj Sano (Okayama, Japan).
I agree with the practice of avoiding circulatory arrest to the brain, doing the whole operation with moderate hypothermia, without deep hypothermia. I use the isolated cerebral and/or myocardial perfusion technique to repair coarctations, arch interruptions, and also to perform the complex DKS operation. I have used this technique in 12 neonates and infants, putting the arterial cannula proximal to the innominate artery, with one death and no neurologic complications. I also use the same technique to do the Norwood procedure, putting the shunt to the innominate artery first and perfusing from the shunt. Brain perfusion is kept from the shunt during the whole procedure.
Recently we have begun cooling the patients only to 25° to 28° C and performing the whole procedure. I believe this technique will have more benefit than using circulatory arrest to the brain.
I have one question. Can you cite any advantages of avoiding circulatory arrest to the brain in your institution or in your patients?
Dr. McElhinney.
Thank you for your comments, Drs. Spray and Sano.
We are well aware of the tremendous contributions that have emerged from the Children's Hospital of Philadelphia in the area of palliation of complex univentricular heart, both for hypoplastic left heart syndrome and for other forms of single ventricle. It is clear that your approach to the present group of patients, namely a Norwood-style arch reconstruction, has a number of advantages. In addition to your outstanding results, I would like to acknowledge the University of Michigan experience reported by Dr. Mosca at this year's Society of Thoracic Surgeon's meeting. Their results with a modified Norwood reconstruction in neonates and infants with single ventricle and subaortic obstruction have been similarly excellent.* We have taken this approach in several patients, and our preference for the modified technique of arch reconstruction with DKS is based on the fact that we can avoid circulatory arrest and eliminate nonnative tissue in the arch repair.
So far, Dr. Spray, we have not had any patients come back with left pulmonary artery or bronchial obstruction after a modified DKS procedure, although we recognize that this is a potential concern when mobilizing the descending aorta and anastomosing it proximally to the ascending aorta or the proximal arch. In addition to this modified DKS repair, we use end-to-side anastomosis of the descending aorta to the ascending aorta or proximal arch in all other cases of coarctation and most cases of interrupted arch. So far we have not observed vascular or airway obstruction as a result of the reconstructed arch in any of our patients, and we believe that extensive mobilization is absolutely essential. We have also not encountered any cases of innominate artery or ascending aortic obstruction resulting either in cannulation or repair of the cannulation site. We use an 8F arterial cannula for these cases, and we would not attempt to cannulate an ascending aorta smaller than 4 mm in diameter.
In response to questions from both Dr. Spray and Dr. Sano, continuous perfusion through the proximally attached modified Blalock-Taussig shunt is a method with which we have no experience, either in this group of patients or in our patients undergoing the Norwood procedure. I do not see any obvious drawbacks to such an approach, and it is an option that we may want to explore in the future. Dr. Sano commented that he uses moderate hypothermia with this approach. Even with our modified DKS technique we continue to use deep hypothermia with low-flow continuous bypass. This is another area that we are likely to examine further as we gain experience with this method.
Regarding Dr. Sano's question about whether we have noticed any advantages of not using circulatory arrest, I would simply note that ours is a fairly small experience and we have not noticed any clear advantages. We have not performed concerted neurologic monitoring in the postoperative period, but none of our patients undergoing the modified DKS procedure has had significant clinical neurologic sequelae. Nonetheless, the literature is clearly available to support avoiding, or at the very least minimizing, circulatory arrest.
Footnotes
From the Divisions of Cardiothoracic Surgerya and Pediatric Cardiology,b University of California, San Francisco, Calif. 
Read at the Seventy-seventh Annual Meeting of The American Association for Thoracic Surgery, Washington, D.C., May 4-7, 1997.
*Mosca RS, Hannein HA, Kulik TJ, et al. Modified Norwood operation for single left ventricle and ventriculoaortic discordance: an improved surgical technique. Ann Thorac Surg. In press.
References
This article has been cited by other articles:
|
|
 |
|
  A. M. Gaca, J. J. Jaggers, L. T. Dudley, and G. S. Bisset III Repair of Congenital Heart Disease: A Primer-Part 1 Radiology, June 1, 2008; 247(3): 617 - 631. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. F Corno and M. Pozzi Safe Innominate Artery Cannulation for Cardiopulmonary Bypass in Neonates Asian Cardiovasc Thorac Ann, December 1, 2007; 15(6): 528 - 530. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. G. Kwak, W.-H. Kim, A. Y. Oh, T. G. Yoon, H.-S. Kim, J. H. Chae, and S. Y. Park Is unilateral brain regional perfusion neurologically safe during congenital aortic arch surgery? Eur. J. Cardiothorac. Surg., November 1, 2007; 32(5): 751 - 755. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Gargiulo, G. Oppido, E. Angeli, and C. P. Napoleone Neonatal aortic arch surgery MMCTS, July 23, 2007; 2007(0723): 2345. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Amir, C. Ramamoorthy, R. K. Riemer, C. R. Davis, F. L. Hanley, and V. M. Reddy Visual light spectroscopy reflects flow-related changes in brain oxygenation during regional low-flow perfusion and deep hypothermic circulatory arrest J. Thorac. Cardiovasc. Surg., December 1, 2006; 132(6): 1307 - 1312. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. L. Hannan, M. A. Ybarra, J. W. Ojito, F. A. Alonso, A. F. Rossi, and R. P. Burke Complex Neonatal Single Ventricle Palliation Using Antegrade Cerebral Perfusion. Ann. Thorac. Surg., October 1, 2006; 82(4): 1278 - 1285. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y.-T. Lan, R.-K. Chang, and H. Laks Outcome of patients with double-inlet left ventricle or tricuspid atresia with transposed great arteries J. Am. Coll. Cardiol., January 7, 2004; 43(1): 113 - 119. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T Rakow and C Bull Same patient, different advice: a study into why doctors vary Arch. Dis. Child., June 1, 2003; 88(6): 497 - 502. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. M. DeCampli, G. Schears, R. Myung, S. Schultz, J. Creed, A. Pastuszko, and D. F. Wilson Tissue oxygen tension during regional low-flow perfusion in neonates J. Thorac. Cardiovasc. Surg., March 1, 2003; 125(3): 472 - 480. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Lim, W.-H. Kim, S.-C. Kim, J.-W. Rhyu, M.-J. Baek, S.-S. Oh, C.-Y. Na, and C. W. Kim Aortic arch reconstruction using regional perfusion without circulatory arrest Eur. J. Cardiothorac. Surg., February 1, 2003; 23(2): 149 - 155. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. M. Bradley, J. M. Simsic, A. M. Atz, and B. H. Dorman The infant with single ventricle and excessive pulmonary blood flow: results of a strategy of pulmonary artery division and shunt Ann. Thorac. Surg., September 1, 2002; 74(3): 805 - 810. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. I. Tchervenkov, S. J. Korkola, D. Shum-Tim, C. Calaritis, E. Laliberte, T. U. Reyes, and J. Lavoie Neonatal aortic arch reconstruction avoiding circulatory arrest and direct arch vessel cannulation Ann. Thorac. Surg., November 1, 2001; 72(5): 1615 - 1620. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C.I. Tchervenkov, S.J. Korkola, and D. Shum-Tim Surgical technique to avoid circulatory arrest and direct arch vessel cannulation during neonatal aortic arch reconstruction Eur. J. Cardiothorac. Surg., May 1, 2001; 19(5): 708 - 710. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. H. Pass, D. E. Solowiejczyk, J. M. Quaegebeur, L. Liberman, K. Altmann, W. M. Gersony, and A. J. Hordof Bulboventricular foramen resection: hemodynamic and electrophysiologic results Ann. Thorac. Surg., April 1, 2001; 71(4): 1251 - 1254. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. A. Pigula, E. M. Nemoto, B. P. Griffith, and R. D. Siewers REGIONAL LOW-FLOW PERFUSION PROVIDES CEREBRAL CIRCULATORY SUPPORT DURING NEONATAL AORTIC ARCH RECONSTRUCTION J. Thorac. Cardiovasc. Surg., February 1, 2000; 119(2): 331 - 339. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. N.K. Odim, H. Laks, D. C. Drinkwater Jr, B. L. George, J. Yun, M. Salem, and V. Allada Staged surgical approach to neonates with aortic obstruction and single-ventricle physiology Ann. Thorac. Surg., September 1, 1999; 68(3): 962 - 967. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Z. Amin, C. L. Backer, C. E. Duffy, and C. Mavroudis Does banding the pulmonary artery affect pulmonary valve function after the Damus-Kaye-Stansel operation? Ann. Thorac. Surg., September 1, 1998; 66(3): 836 - 841. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
