| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2007;84:612-618
© 2007 The Society of Thoracic Surgeons
a Division of Cardiothoracic Surgery, Department of Surgery, Emory University School of Medicine, Atlanta, Georgia
b Children’s Healthcare of Atlanta at Egleston, Division of Pediatric Cardiology, Emory University School of Medicine, Atlanta, Georgia
Accepted for publication March 7, 2007.
* Address correspondence to Dr Kanter, Division of Cardiothoracic Surgery, Emory University School of Medicine, 1365 Clifton Rd, Atlanta, GA 30322 (Email: kkanter{at}emory.edu).
Presented at the Fifty-third Annual Meeting of the Southern Thoracic Surgical Association, Tucson, AZ, Nov 8–11, 2006.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
Methods: Three approaches were used to manage 36 consecutive infants with coarctation and VSD. Group I had staged coarctation repair with or without pulmonary artery banding, followed by VSD closure with two separate operations (two-stage, n = 11); Group II had coarctation repair and VSD closure on cardiopulmonary bypass (CPB) with circulatory arrest or regional perfusion during coarctation repair (one-stage, one-incision, n = 10); Group III had coarctation repair without CPB through a thoracotomy, followed by VSD closure during the same operation (one-stage, two-incisions, n = 15).
Results: No patients died. One recoarctation occurred in group II. Group II had significantly longer times for CPB (135.6 ± 31.8 versus 94.3 ± 29.8 minutes for group I; 67.6 ± 16.7 minutes for group III; p < 0.001) and combined regional perfusion/circulatory arrest (30.0 ± 17.0 versus 5.3 ± 11.9 minutes for group I, 1.1 ± 4.4 minutes for group III, p < 0.0001). Group III compared with group II had significantly shorter lengths of stay in the intensive care unit (119.5 ± 64.8 versus 220.8 ± 198.8 hours, p = 0.04) and hospital (8.4 ± 3.8 versus 24.4 ± 24.4 days, p = 0.01). Combining values for the two hospitalizations in the group I infants, lengths of stay in the intensive care unit (178.8 ± 70.8 hours) and hospital (20.5 ± 11.6 days) were intermediate between groups II and III.
Conclusions: Primary repair of infants with coarctation and VSD using a one-stage approach through separate incisions affords excellent clinical results. One can avoid prolonged aortic cross-clamping, CPB, and circulatory arrest/regional perfusion. Compared with the group undergoing combined coarctation and VSD repair simultaneously by sternotomy, total lengths of stay in the intensive care unit and hospital were significantly decreased.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
Lately, there has been enthusiasm for a one-stage, one-incision approach in which both the coarctation and the VSD are repaired at the same operation through a sternotomy involving the use of CPB. Typically, the coarctation is repaired during a period of deep hypothermic circulatory arrest (DHCA) or with the use of regional cerebral perfusion. The VSD is then closed using standard CPB techniques. The one-stage, one-incision technique has the advantages of complete repair of all the cardiac defects at one operation without interim palliation. This approach also allows ready access for correction of proximal aortic arch hypoplasia, if present. The one-stage, one-incision strategy has the disadvantages of the use of circulatory arrest or regional cerebral perfusion in the infant. Also, this is clearly a more complex operation than simple coarctation repair with or without pulmonary artery banding.
Although both operations have their proponents, we have questioned, what is the best approach for this common combination of defects? We believed it would be advantageous to attempt to incorporate the advantages of both the two-stage approach and the one-stage, one-incision approach. We therefore have applied the strategy of a one-stage technique of repairing both the coarctation and the VSD during the same operation but with two separate incisions. The one-stage, two-incision technique involves repair of the coarctation through a left thoracotomy, followed by VSD closure through a sternotomy under the same anesthetic. This technique has the potential advantages of repair of all the defects in infancy, avoids palliation and the need for circulatory arrest or regional cerebral perfusion for the coarctation repair, and, putatively, results in a much simpler open heart operation compared with the one-stage, one-incision approach.
We have reviewed our experience with these three surgical approaches in 36 consecutive, nonrandomized infants presenting with coarctation of the aorta and a hemodynamically significant VSD treated at our institution from 2002 to 2006.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
Patient Population
From 2002 to 2006, 36 consecutive nonrandomized infants (age younger than 90 days) presenting with coarctation of the aorta and a hemodynamically significant VSD were operated on at our institution. The patients were divided into three groups defined by the surgical strategy at the discretion of the attending surgeon. The distribution of operative strategy among the three groups for each year is shown in Fig 1. As time went on and experience was gained, the tendency was to use the one-stage, two-incision technique. Nonetheless, one can see from Fig 1 that each of the techniques was applied every year in this study, except for 2003. Patient characteristics at the time of initial operation are summarized in Table 1. The three groups were comparable with respect to age, weight, and need for prostaglandin infusion at the time of the first operation.
|
|
Operative Techniques
The two-stage group included 11 infants who had aortic coarctation repair through a left thoracotomy without closure of the ventricular septal defect, and 7 had concomitant pulmonary artery banding at this time. The coarctation was typically repaired by resection with end-to-end or extended end-to-end anastomosis. Because of distal arch hypoplasia, 4 (36%) infants also had reverse subclavian flap repair of the hypoplastic aortic arch, as described previously by our group [2]. After a mean interval of 5.1 ± 3.2 months (range, 12 days to 9.8 months), the VSD was closed and the pulmonary artery band was removed if it had been placed at the time of coarctation repair. Of interest was that 2 of the 11 children in this two-stage group had VSD closure as the first operation, followed by coarctation repair within 2 weeks owing to unrecognized or underestimated coarctation at the time of the initial presentation.
The second group of 10 patients underwent simultaneous primary repair of both the aortic coarctation and the VSD through the same sternotomy incision (one-stage, one-incision group). The coarctation was repaired using an extended end-to-end anastomosis in 6 patients or homograft patch enlargement of the coarctation and aortic arch in the remaining 4. During the coarctation repair, deep hypothermic circulatory arrest was used in 4 infants, regional cerebral perfusion as modified from the technique of Pigula et al [3] was used in 4, and a combination of circulatory arrest and regional perfusion was used in the remaining 2 patients.
The third group (n = 15) had repair of both the coarctation and VSD during the same operation in infancy using two separate incisions (one-stage, two-incision group). First, the infant underwent a left thoracotomy with resection of the aortic coarctation with end-to-end or extended end-to-end anastomosis. Because of distal aortic arch hypoplasia, a reverse subclavian flap repair of the hypoplastic arch was performed in 5 (33%) of the 15 patients. The incision was then closed with a chest drainage catheter in place. The patient was repositioned and through a median sternotomy underwent standard Dacron (DuPont, Wilmington, DE) patch closure of the ventricular septal defect using moderate hypothermia with standard perfusion techniques, without the need for profound hypothermia with circulatory arrest or regional perfusion.
All patients were followed up by their referring cardiologists with echocardiograms, as indicated. The postoperative echocardiograms were assessed for recurrent coarctation or residual arch obstruction by a single observer (WTM) blinded to the operative group.
Statistical Methods
Continuous variables are presented as mean ± standard deviation. Comparison between groups for continuous variables was performed using analysis of variance. Categoric values were compared by 
2 analysis. Statistical differences were considered significant at p < 0.05.
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
|
|
|
|
|
Comment |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
A more recent report from Isomatsu and colleagues [6] in Tokyo reviewed 79 infants younger than 3 months old with coarctation and a hemodynamically significant VSD. All infants underwent subclavian flap repair of the coarctation with a concomitant pulmonary artery band. The VSD was closed after an interval of 10.4 ± 9.6 months with removal of the pulmonary artery band. Early mortality was 2.5% in this group, with a 10-year actuarial survival of 92.3%. Of importance was the 90.4% freedom from recoarctation at 10 years [6].
An interesting variation on the two-stage approach to infants with coarctation and a VSD was reported by Bonnet and colleagues from Paris [7]. They described 11 infants younger than 30 days old with coarctation and a medium-sized VSD that was nonrestrictive but measured a mean 5.0 ± 0.7 mm (range, 4 to 7 mm). They performed an extended end-to-end repair of the aortic coarctation and placed an absorbable pulmonary artery band using polydioxanone. There were no early or late deaths. Only one recoarctation occurred, which was successfully balloon dilated. Of note, complete spontaneous closure of the VSD occurred in 4 patients, and 6 had partial closure with no significant residual left-to-right shunt. The pulmonary artery band was completely reabsorbed in all patients by 6 months without any residual pulmonary artery stenosis or distortion. Only 1 patient with a trabecular VSD required further surgical intervention, giving a 91% freedom from reoperation in this series.
Looking at a published series of patients undergoing a one-stage, one-incision repair, Gaynor and colleagues [8] from the Children’s Hospital of Philadelphia presented 25 infants younger than 3 months old with coarctation and a hemodynamically significant VSD. Using a period of circulatory arrest, they repaired the coarctation and arch either primarily in 4 or with a patch in 21. Their average circulatory arrest time was 38 ± 12 minutes. Early mortality was 4%, with no late deaths. Of importance, however, 7 of the 25 patients required eight reinterventions for coarctation, giving a 75% freedom from recoarctation at 2 years.
Brouwer and colleagues [9] from the Netherlands reported 80 infants younger than 3 months old with coarctation and VSD. They performed a two-stage repair in 64 (80%) of these 80 infants in whom only 10 had pulmonary artery banding. Sixteen children (20%) had a one-stage, one-incision repair using hypothermic circulatory arrest during the repair of the coarctation. The early mortality in the two-stage group was 4.7% compared with an 18.8% early mortality in the one-stage, one-incision group, although the mortality difference did not achieve statistical significance. Of interest was 41% of the two-stage patients did not require further operation because their VSD closed completely or became hemodynamically insignificant with time. The freedom from recoarctation at 5 years in the two-stage group was 91.3% compared with a freedom from recoarctation in the one-stage group of only 60% (p = 0.018).
With these studies in mind, we searched for a compromise between the two-stage approach and the one-stage, one-incision approach hoping to achieve the benefits of each while avoiding the disadvantages. One advantage of the two-stage repair is that it affords a fairly straightforward repair of the coarctation even if there is a somewhat hypoplastic arch, as was seen in our series. The avoidance of CPB and circulatory arrest or regional cerebral perfusion in the first stage makes for a simpler operation in these infants. At the second stage, when the child is older and bigger, the VSD closure is arguably more straightforward compared with neonatal VSD closure. The disadvantage of the two-stage repair is the need for a second hospitalization and operation as well as the potential detrimental effects of a pulmonary artery band with pulmonary artery distortion and the need for pulmonary artery reconstruction if a band had been placed. The patient also has a period of palliation between the two operations.
The one-stage, one-incision approach has the advantages of complete correction of all defects in infancy without the need for palliation. The disadvantage, of course, is that it is a much bigger operation with the unwanted necessity for the use of hypothermic circulatory arrest or regional perfusion for the coarctation repair, and as noted, there is the published increased incidence of recoarctation using this strategy.
We have used a third option for this group of infants using simultaneous repair in infancy of both defects through two separate incisions. In our series, this has resulted in excellent clinical results comparable with the other two groups, with no deaths either early or late and no episodes of clinically important recoarctation. The operative variables of cross-clamp time and CPB time were significantly longer in the one-stage, one-incision group compared with the other two groups (Table 2, Fig 2). The one-stage, two-incision strategy allowed us to avoid the use of DHCA or regional cerebral perfusion in all but 1 of our 15 patients.
Although some have maintained that limited periods of properly performed circulatory arrest or regional perfusion with adequate cooling are not injurious to the brain, we know that prolonged periods are detrimental [10]. A recent study from the Children’s Hospital of Philadelphia showed that increasing the duration of DHCA was associated with postoperative seizures after neonatal heart surgery, a putative marker for brain injury [11]. As with most biologic systems, there must be a spectrum of safety. We believe it is reasonable, therefore, to maintain that if a long period of circulatory arrest is injurious, then a more limited period can not be entirely without risk, particularly since that magical exact time cutoff for safety with circulatory arrest is elusive [12]. We therefore contend that a surgical strategy that allows avoidance of circulatory arrest or regional cerebral perfusion without compromising the quality of the surgical repair, as is seen in our one-stage, two-incision group, is valuable and worth pursuing.
Survival and rates of recoarctation were equally good among the three groups, but the use of hospital resources was significantly different. Delayed sternal closure was much more frequent in the one-stage, one-incision group (60%) compared with the other two groups (Table 2). The ICU and hospital lengths of stay were significantly shorter in the one-stage, two-incision group compared with the one-stage, one-incision group (Table 2, Fig 3 and Fig 4). When the durations of hospital stay for both hospitalizations in the two-stage group are combined, this group had a significantly longer total duration of hospitalization compared with the one-stage, two-incision group (Table 2, Fig 4). We believe that this one-stage, two-incision approach presents the advantages of both of the other more time-tested techniques without their disadvantages. Specifically, the heart defects are totally repaired during one operation in infancy without palliation and without the routine use of circulatory arrest or regional perfusion. The coarctation repair is relatively straightforward through a thoracotomy, with no recoarctation in our series. The improved use of hospital resources as measured by less use of delayed sternal closure and shorter ICU and hospital lengths of stay cannot be overlooked.
We, therefore, recommend the one-stage, two-incision approach for infants with coarctation of the aorta and a hemodynamically significant VSD. Other strategies may be more appropriate in certain groups of patients. For example, for the infant with coarctation and multiple VSDs or a Swiss cheese septum, we would recommend a two-stage approach with coarctation repair and pulmonary artery banding in infancy, followed by VSD(s) closure and removal of the band at a later date. Also, for the infant with associated significant hypoplasia of the proximal aortic arch (between the innominate artery and the left carotid artery), we would recommend a one-stage, one-incision approach because it is difficult to successfully deal with significant proximal arch hypoplasia through a thoracotomy incision without the aid of CPB.
This study has several important limitations. First, there are the usual limitations of a retrospective study with relatively small sample sizes. Second, because these patients were not randomized to the specific surgical strategy used, we cannot eliminate the possibility of a selection bias. Although our retrospective analysis of preoperative aortic arch sizes suggests that the three groups were not significantly different (Table 1), we can not rule out the presence of some subtle, unmeasured selection bias which we have not identified which could skew the results of this study. Only a prospective randomized trial can answer this question satisfactorily, which is beyond the scope of this report.
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
A couple of things struck me in your data, however. One is that you included patients that were up to 3 months of age. So, we are dealing with, I assume, babies born with ductal-dependent coarctations as well as patients who were home and were not ductal-dependent. Therefore, I am not quite certain that we are dealing with the same group of patients throughout the study period.
Also, it should be emphasized that this study is nonrandomized and the choice of technique was at the discretion of the surgeon. One cannot help but wonder what subtle, perhaps nonmeasurable, findings may have influenced you or your colleagues to approach your patients one way or the other.
Now, I am not a statistician but I would suggest that had I brought these data to my statistician at home, he or she would probably castigate me for using statistical data in a retrospective analysis, particularly one with relatively small numbers. I wonder if you could comment about that, or if you can justify your choice of statistical methods in the data that you presented?
A few questions come to mind. First, could you explain the excessively long hospital and ICU stays in your group II patients? I was struck in reviewing the abstract that the patients who had one incision and one operation were in the hospital nearly 1 month and in the intensive care unit a good week. That is out of line to what I think the typical patient would be in our own center. Although there are always outliers, statistically you demonstrated that these patients had a very long stay.
Second, have you seen neurologic injury in any of these patients? The point was made that the amount of cardiopulmonary bypass time and the amount of circulatory arrest or regional cerebral perfusion times are statistically lower with the one-stage but two-incision approach. However, I think one would have to say that there are no data that I am aware of to indicate that a relatively brief period of circulatory arrest, when used appropriately, is associated with neurologic damage. I would add that neurologic damage associated with a properly conducted circulatory arrest of 15 or 20 minutes, which is typically what it would take to repair the coarctation, would not leave one with a scar on the brain, and I wonder if you have any data to suggest otherwise?
Finally, do you feel that you may have selected the more long and difficult anatomic coarctations for one-stage repairs versus the two-stage repairs? Thank you very much for the opportunity to discuss the paper.
DR KANTER: Thank you for your comments. With regards to age, we did take the patients out to 90 days of age, but the average age was about 2 weeks. The median age, which is probably more descriptive, was similar in all three groups and ranged from 5 days to 13 days, so I think the ages were similar.
In choosing which surgical strategy to employ, I think that there had to be some bias involved; for example, with the surgeon looking at the arch and saying, "this is an arch which would be much easier to deal with from the front." For that reason, we did the retrospective blinded evaluation of the preoperative echoes to measure if the arches were different. I expected them to be different, because it was my impression that we were selecting the more difficult arches to have the one-incision, one-stage operation, but was surprised to find that the arches really weren’t much different. I think that there was some subtle decision-making. Certainly, the 1.3-kg child was much more easily repaired through one incision.
I can’t comment on the statistics. I don’t think the paper would have been accepted if it had no statistics in it, so I felt obligated to put them in. You can interpret them as you wish.
As far as our long hospital stays are concerned, it seems like in Georgia we have a lot of trouble with feeding in children who have had big operations. Most of these children who were in the hospital for an average of over 3 weeks were not there for cardiac reasons but because they had feeding difficulties. We see this with our Norwood patients as well. When one dissects the arch and coarctation from the front, I wonder if the vagus nerve and the recurrent laryngeal nerve are bruised, thus contributing to postoperative feeding and swallowing difficulties. It is much easier to protect those nerves when you are doing the arch or coarctation repair from the side. That is just speculation.
Finally, I have no hard data to demonstrate differences in neurologic outcome. I agree with you that a well conducted operation with adequate cooling and less than 30 minutes arrest is probably safe, but you just never know. It is a biologic system, so there must be a continuum of brain injury with length of circulatory arrest rather than an all-or-none phenomenon.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. Talwar, S. Kumar Choudhary, and B. Airan Single-stage versus 2-stage repair of coarctation of the aorta with ventricular septal defect. J. Thorac. Cardiovasc. Surg., November 1, 2008; 136(5): 1390 - 1391. [Full Text] [PDF]
|
|
|
|
|
|
H. L. Walters III, C. E. Ionan, R. L. Thomas, and R. E. Delius Single-stage versus 2-stage repair of coarctation of the aorta with ventricular septal defect. J. Thorac. Cardiovasc. Surg., April 1, 2008; 135(4): 754 - 761. [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 |
