| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Section of Cardiothoracic Surgery, James W. Riley Hospital for Children, and Indiana University School of Medicine, Indianapolis, Indiana
b Section of Pediatric Cardiology, James W. Riley Hospital for Children, and Indiana University School of Medicine, Indianapolis, Indiana
Accepted for publication July 15, 2009.
* Address correspondence to Dr Brown, Section of Cardiothoracic Surgery, Indiana University School of Medicine, 545 Barnhill Dr, EH 215, Indianapolis, IN 46202-5123 (Email: jobrown{at}iupui.edu).
Presented at the Forty-fourth Annual Meeting of The Society of Thoracic Surgeons, Fort Lauderdale, FL, Jan 28–30, 2008.
| PEDIATRIC CARDIAC SURGERY:
The Annals of Thoracic Surgery CME Program is located online at http://cme.ctsnetjournals.org. To take the CME activity related to this article, you must have either an STS member or an individual non-member subscription to the journal.
|
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Methods: We retrospectively reviewed our experience with 1,012 patients undergoing initial repair of coarctation between 1960 and 2008. During that time, 103 patients (10%) required reintervention. Median age at reintervention was 6.5 years (range, 2 weeks to 44 years) and median weight was 12 kg (range, 1.9 to 94 kg). Fifty-nine patients with recoarctation had surgical repair, and 44 patients were treated with balloon aortoplasty with or without stent placement.
Results: Ninety-five percent of patients have been followed up (median time, 14.2 years; range, 2 months to 42 years). There were 5 late deaths. Actuarial survival was 98% at 15 and 40 years in patients with surgical reintervention, and it was 91% (p = 0.001) at 15 years in patients with balloon aortoplasty reintervention. A second redo coarctation of the aorta reintervention was performed in 12 patients: 8 patients after percutaneous intervention (nonsurgical) and 4 patients after surgical recoarctation repair. The median interval between first and second reintervention was 3.5 years (range, 1 month to 14 years). One patient who had two dilations underwent a third and fourth reintervention: patch enlargement and pseudoaneurysm resection. Freedom from reintervention in the surgical group was 96% at 15 years and 94% at 40 years, which was compared with actuarial freedom from reintervention for patients with percutaneous intervention (balloon/stent) at 15 years (82%; p < 0.001).
Conclusions: Our study demonstrates that surgical repair of recurrent coarctation of the aorta can be performed safely and with excellent results. The recurrence after surgical reintervention is low, and most patients to date have not required further intervention. Balloon aortoplasty as an alternative method of managing recoarctation is efficient and less invasive than surgery; however, well-described complications may occur. Recurrence rates with angioplasty are significantly higher than with surgery.
Coarctation of the aorta (CoA) is the fifth most common cardiovascular anomaly requiring surgical intervention in infants and children. The incidence of CoA is 4 in 10,000 live births, accounting for 5% to 8% of children with congenital heart disease [1]. The natural history of this congenital disorder, through progressive hypertension, leads to heart failure and premature death [2]. Without correction, the mean life expectancy of patients with aortic coarctation is 35 years, and 90% die before reaching age 50 years of age [3]. As the disease progresses, various vascular complications can occur, such as coronary disease, aortic dissection and rupture, hemorrhagic cerebrovascular accident, or bacterial endocarditis [3, 5–8].
Since surgical repair of CoA became available in 1944 [4], great progress has been made in the diagnosis and treatment of CoA. Perioperative morbidity and mortality have been greatly reduced, particularly in cases of uncomplicated CoA not associated with complex congenital heart disease [5–8].
Persistent arterial hypertension after repair is performed at an older age, and recurrence of CoA in repairs done in neonates, however, continues to challenge our management strategies [5, 6]. Postoperative hypertension, which occurs in 7% to 33% of patients, may be a result of recurrent or residual CoA or may be idiopathic [8, 9]. The incidence of postoperative recurrent CoA ranges from 5% to 50%, depending on the criteria and methods used in making the diagnosis [9]. The rate of recurrence varies significantly, depending on the age at initial repair [9, 10]. The specific surgical method of repair, suture material used, and sewing technique do not appear to be as important as once suggested in determining recurrence [9].
Balloon aortoplasty with or without stent placement of CoA has been introduced in the last several years as an alternative method of managing aortic coarctation, particularly recurrent CoA. This is certainly a less invasive method of intervention, although well-described complications may occur and recurrence rates are relatively high (more than 50% for primary treatment in infants) [11–13]. The role of balloon aortoplasty and stent placement for recurrent CoA needs to be further defined. We report our experience with the management of aortic recoarctation treated by balloon aortoplasty/stent placement or repeat surgery and report the outcomes of both approaches at our institution.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Study Subjects
Between January 1960 and January 2008, 1,012 patients with isolated CoA or with complex coarctation, or both, underwent repair at the James Whitcomb Riley Hospital for Children in Indianapolis, Indiana. There were 648 males (64%) and 364 females (36%). Diagnosis included isolated CoA in 506 of 1,012 patients (50%), CoA and ventricular septal defect (VSD) in 213 (21%), and CoA with complex intracardiac anomalies (complex coarctation) in 293 (29%). The distributions of associated cardiovascular anomalies and noncardiac syndromes are shown in Table 1. Associated isthmus hypoplasia was defined as an isthmus diameter of less than 40% of the diameter of the ascending aorta [14]. Arch hypoplasia was defined as a proximal or distal transverse arch diameter of less than 50% of the diameter of the ascending aorta [14]. Isthmus hypoplasia or arch hypoplasia, or both, was initially present in 314 (31%) of our 1,012 patients. Bicuspid aortic valve was present in 587 patients (58%), and congenital syndromes (Turner, Down, and so forth; all syndromes are given in Table 1) were present in 202 patients (20%).
|
|
|
Data Acquisition
A retrospective review of medical records was performed with regard to initial cardiac diagnosis, pathophysiologic findings, surgical treatment, and hospital mortality. Data from outpatient visits and from patients who died after hospital discharge were obtained from physician, hospital records, or death certificates. Medical records and clinical charts were reviewed for all CoA patients and included operative records as well as preoperative and postoperative catheterization and echocardiography data. Data collected from the operative admission included diagnosis, previous operative procedures, age, sex, and weight at operation. Patient follow-up data were obtained from hospital and clinic visit records, and were retrieved retrospectively from electronic chart review.
After the initial CoA repair, the patients routinely returned to the clinic for measurement of resting blood pressures and echocardiograms. Resting arm-leg gradients were also obtained and were considered abnormal if greater than 20 mm Hg. Exercise arm-leg gradients were also obtained in some patients and, if more than 40 mm Hg, were likewise considered abnormal. The patients who had persistent hypertension, elevated arm-leg gradients, or significant pressure gradients on echocardiogram (more than 30 mm Hg, or more than 20 mm Hg in case of inadequate collaterals) underwent echocardiography or catheterization, or both. The patients with re-CoA were sometimes discussed with cardiothoracic surgeons before a decision regarding the type of reintervention was made. At other times, the interventional cardiologists proceeded with balloon dilation and stenting if they felt it was feasible and did not discuss it with surgeons.
Postintervention aneurysm formation was defined according to the criteria of Beekman and coworkers [15]. These included either a fusiform dilation at the CoA site with a diameter greater than 150% of the aortic diameter at the diaphragm or a discrete saccular dilation that was not present at the preangioplasty imaging study. If an aneurysm was detected on a follow-up aortogram, the preintervention aortogram was reviewed to ascertain if this irregularity of the aorta was present before treatment.
Surgical Technique
The surgical technique for both primary and reoperative re-CoA treatment has been described previously [8]. Before 1993, all patients presenting with re-CoA underwent redo surgery. More recently, balloon aortoplasty and stent insertion were introduced; some recurrences are preferentially managed by means of transcutaneous balloon aortoplasty or stent insertion (Fig 2).
|
A number of surgical techniques were employed in the 59 patients who underwent surgical revision (Fig 3): incision of the restricted area and patch enlargement with PTFE or Hemashield was performed in 43 patients, 7 patients had tube graft (PTFE) interposition, 2 patients had subclavian flap angioplasty, and 1 had an extended end-to-end repair. Of the 6 patients with aneurysm/pseudoaneurysm formation, 2 underwent pseudoaneurysm resection and patch repair, and 4 patients underwent aneurysm resection and tube graft interposition. Of the 44 patients who had first nonsurgical repair, balloon aortoplasty alone was employed in 31, and 13 had concomitant stent placement (Fig 3).
|
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
A total of 117 reinterventions were required for re-CoA repair in 103 patients, including surgical treatment (n = 69), transcutaneous balloon angioplasty (n = 44), and stent placement (n = 14). Among these patients, 91 had one reintervention, 11 had two reinterventions, and 1 patient had four reinterventions. The most common indication for the first re-CoA repair was recurrent aortic obstruction at the site of previous CoA repair, which was present in 94 (91%) of the re-CoA patients. A less-frequent primary indication for re-CoA repair was aneurysm formation, which occurred in 9 patients (9%). Seven aneurysms occurred after the initial patch aortoplasty, coarctation repair, and 2 occurred after reintervention balloon aortoplasty and stent placement (Table 3). The median time interval between primary CoA repair and redo surgery was 7.5 years (range, 1 week to 33 years), and the median age at reoperation was 9 years (range, 2 weeks to 44 years). The median time between initial CoA repair and balloon aortoplasty was 1 year (mean, 5.4 ± 6.3; range, 2 weeks to 15 years). Twenty-six patients (26 of 44; 60%) had their reintervention during the first year after the initial CoA surgery. The median age at balloon aortoplasty reintervention was 2 years (range, 2 months to 22 years). The option to return to the operating room for surgical repair of the recurrence after 1993, when balloon aortoplasty was available, was based the anatomic location, severity of the recurrence, and age of the patient. As a general rule, the younger the patient, the more severe and proximal the obstruction, the more likely the patient was referred for surgical reintervention.
|
Patient Survival
There were no intraoperative deaths after surgical reintervention. There was 1 late death of unknown cause (2%; 1 of 59) 2 years after re-CoA surgery in a patient with a single ventricle. There were no early deaths among the 44 patients who had balloon aortoplasty or stent placement, but there were 4 late deaths (9%; 4 of 44) from 6 months to 3 years (mean, 1.4 ± 1.2 years) after aortic balloon aortoplasty (Table 3). All late deaths were unrelated to balloon reintervention. Two deaths were in patients with single-ventricle physiology in whom heart failure developed. One patient with Shone's anomaly underwent mitral valve replacement 2 years after angioplasty and died 1 year after mitral valve replacement. The last late death was of a patient who underwent balloon aortoplasty of the recoarctation site and died 9 months later of multiorgan failure when he had VSD closure and pulmonary artery patch arterioplasty.
Overall survival estimated by the Kaplan-Meier method was 97% at 1 year, and 95% at 5, 10, 15, 20, and 40 years (Fig 4A). Actuarial survival was 98% at 5, 15, and 40 years among patients with surgical reintervention, and 91% at 15 years among patients with balloon aortoplasty reintervention (p = 0.001). Univariate and multivariate analyses showed that the presence of complex intracardiac anomalies (p < 0.001) and age less than 1 year (p < 0.003) were significant predictors of mortality after re-CoA repair.
|
Indications for second reintervention were residual obstruction at the site of previous CoA and re-CoA repair (n = 9), migration of a stent over the orifice of an aberrant right subclavian artery (n = 1), aneurysm of distal aortic arch involving the origin of the left subclavian artery (n = 1), and resection and graft replacement of an infected pseudoaneurysm at the site of stent placement (n = 1).
The overall freedom from re-CoA reintervention, as estimated by the Kaplan-Meier method, was 94% at 10 years, 89% at 15 years, and 88% at 20 and 40 years (Fig 4B). Freedom from reintervention in the surgical group was 96% at 15 years and 94% at 40 years as compared with the actuarial freedom from reintervention for patients with percutaneous intervention (balloon/stent) at 15 years of (82%; p < 0.001).
Echocardiographic Gradients
The peak gradients across the coarctation site (preinterventional, postinterventional, and at last follow-up) for the surgical and balloon aortoplasty groups are shown in Figure 5. Fifteen patients had a peak gradient of more than 30 mm Hg at last follow-up (4 patients after surgical treatment and 11 patients after balloon/stent group). Three of them (surgical group) have been treated with reintervention: 1 patient with graft insertion underwent patch enlargement, and 2 patients with patch enlargement underwent balloon aortoplasty and stent insertion (Table 3); 1 remaining patient is awaiting reintervention. In the balloon/stent group of patients, 8 have been treated with a second reintervention (six surgical procedures and two additional balloon aortoplasty), and 3 other patients are awaiting reintervention (Fig 3).
|
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Aortic recoarctation was diagnosed in our study when, after the initial discharge of a patient who had undergone CoA repair with a good result developed a resting arm–leg blood pressure gradient of 20 mm Hg or greater or who had a significant Doppler echocardiographic gradient. Reintervention was elected when signs of systemic hypertension were present or when echocardiography demonstrated hypertrophy or a dilated left ventricle. A small group underwent reintervention if there was an aneurysm at the original repair site.
The primary objective of this study was to compare the outcome of surgical intervention to the more recently introduced technique of balloon aortoplasty with or without stent placement. Balloon aortoplasty is efficient, less invasive, and can treat recoarctation at a lower initial cost and shorter hospital stay. Balloon aortoplasty became the first option of treatment for re-CoA at our institution after 1993, as shown in Figure 1. Very little has been published comparing the outcomes of these two approaches. We undertook this retrospective analysis of these two approaches to recoarctation to see if there was a difference in outcomes. Zoghbi and associates [5] has recently reported a large series of 97 patients with re-CoA intervention and found that balloon aortoplasty had more frequent major complications and a higher recurrence rate. Our analysis supports their conclusions, and is shown in Figure 5 and Table 3. Major complications were three to four times as frequent with balloon aortoplasty as they were with surgery in the Zoghbi study [5] and in our review (Table 3).
Careful preoperative investigations including angiography, magnetic resonance image scanning, and echocardiography may be necessary to allow the best choice of appropriate reintervention therapy. If the site of aortic obstruction is proximal to the CoA repair site and if the transverse aortic arch remains significantly hypoplastic, the lesion is preferentially approached through a median sternotomy. If the recurrent obstruction is just proximal, at or distal to the left subclavian artery, the lesion may best be approached through a redo left thoracotomy.
It is interesting that of the 314 patients (31%) who presented with isthmus or transverse aortic arch hypoplasia at initial coarctation presentation, only 3 patients have required subsequent transverse arch patch augmentation to date. This lack of the need to treat transverse arch hypoplasia demonstrates that the transverse arch will grow in most young infants if the isthmus and coarctation are adequately treated at initial presentation.
When comparing surgical intervention and balloon aortoplasty, results do not show any statistical difference in early and late mortality between the two methods. Conversely, analysis of early and late morbidity seems to demonstrate that surgical repair is safer, more effective, and has a lower incidence of persistent obstruction, and this trend has been demonstrated by others [7]. In particular, no neurologic complications such as spinal cord ischemia, phrenic nerve injury, or recurrent nerve paralysis were encountered in our series and in another surgical series [17] when descending aorta perfusion or hypothermia were utilized. Morbidity after balloon aortoplasty is not infrequent. The most frequent complication reported after balloon aortoplasty are cerebrovascular accidents and femoral artery injuries [18]. Neither of these complications was observed in our study.
Although the present data seem to favor redo surgery in most patients with recurrent aortic obstruction, some authors prefer intravascular stents in their older patients for this indication and have shown good short-term results [19, 20], but long-term outcome is unknown. Aneurysm formation has varied from 0% to 14% and emphasizes the need for long-term follow-up. Other complications of stent implantation after recoarctation have included femoral artery trauma, neurologic damage, postcoarctation syndrome, and stent migration [12, 13, 18, 21].
Although no patient in our aortoplasty series required surgery for femoral artery trauma, the loss of a femoral pulse was not routinely recorded. Two patients (4.2%) of the aortoplasty group and 14% of the 14 stented patients had aneurysms at the site of stent placement. In 1, the aneurysm was acutely infected and had a contained rupture and required emergent surgery. One patient had stent migration over the subclavian artery, and 6 patients required surgery for incomplete relief of their gradients (Table 3 and Fig 3).
Aortic aneurysms develop in as many as 14% of patients after balloon aortoplasty in some series and are less frequent after surgical reintervention. Aneurysm formation after primary patch aortoplasty has been reported extensively, with the reported incidence varying between 2% and 51% [14, 22]. The cause of aneurysm formation after patch aortoplasty has been attributed to several different factors: patch material, resection of CoA ridge, age, and aortic arch hypoplasia [14, 22–26]. Aneurysm formation was encountered in 9 of our patients (1%) of 980 surviving primary CoA repair, 7 of 211 (3.3%) after initial patch aortoplasty, and 2 of 44 (4.6%) after balloon aortoplasty or stent implantation. Most of the aneurysms after primary prosthetic patch repair in our series developed after procedures involving CoA ridge resection and prosthetic patch placement performed before 1988. Neither age nor transverse arch-isthmus hypoplasia were predictors of late aneurysm formation in our series.
In summary, we have demonstrated that surgical repair of recurrent obstruction and aneurysm formation at the site of primary CoA repair produces good and lasting results in the majority of patients and remains the "gold standard" of reintervention. Balloon aortoplasty with or without stenting, with or without covered stents, has become the initial procedure of choice at most institutions including our own, even though one third of patients have complications or persistent obstruction and may eventually require surgery. Transverse aortic arch obstruction, although uncommon, should almost always be approached surgically through sternotomy.
Surgery for re-CoA after an attempted balloon aortoplasty with or without stent is theoretically more treacherous for the patients than initial redo surgery. The bridging collaterals that formed during the recurrence may regress rapidly after balloon aortoplasty and may result in a lower distal aortic pressure during surgery and clamping of the aorta. Left-side heart bypass or femoral artery/femoral vein bypass is more likely to be required under these circumstances because the distal aortic pressures falls below the desired level of 40 to 50 mm Hg or the evoked potentials may diminish. Both patients who had aneurysms after stent placement required left-side heart bypass. If a stent was utilized and failed to adequately treat the recurrence, then that segment of the aorta will likely require resection and graft interposition as the stent can rarely be removed without destroying the integrity of the aortic wall. Stent failure may increase the necessity to do an extra-anatomic bypass of the recurrent CoA. The incidence of recurrence is decreasing in most centers. Patients who have recurrence will require long-term follow-up, and the methodology to address recurrence will continue to evolve.
It is conservative to say that most patients with severe long-segment recoarctation are best treated with redo surgery. Patients with less severe, shorter segment re-CoAs should have an initial trial of balloon aortoplasty. Stent placement should be reserved for teenagers and adults in whom continued growth of the aorta is not desired or expected. Prevention of recurrence will continue to be the goal of surgery for primary coarctation.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
DR BROWN: At the time of initial coarctation surgery we close the pleura over the repair. This facilitates a redo greatly. We routinely circumferentially dissect out the aorta to gain full proximal and distal control. I've never felt very secure with a side-biter in redo coarctation surgery.
DR DOMINIQUE R. METRAS (Marseille, France): Looking at your conclusions, does this mean that you recommend surgery for all recoarctations, or if not, how do you choose to have dilatation or surgery? How do you make the decision?
DR BROWN: Your last statement was correct. We rarely discuss the therapeutic option until after the cardiologist have already attempted balloon angioplasty. Basically we only get referred after their unsuccessful attempts or complications. The cardiologists see the patients first. Although we discuss all of the patients requiring surgery with our cardiologists ahead of time, our cardiologists rarely discuss their modes of therapy with us before they intervene. I doubt that our situation is much different than your experience in your institution.
DR CARL L. BACKER (Chicago, IL): John, I believe you previously published your surgical results with recoarctation more than 10 years ago. You concluded that reoperation was the gold standard for recoarctation then, and I think you've confirmed that it is still the gold standard. I have two questions.
In some patients you used an interposition graft and in some patients a Dacron patch. I have to admit that in patients that have a recoarctation that are adult size I have switched from using the patch technique to an interposition graft for fear of aneurysm formation. I noticed you did not have any late aneurysms, but other people have reported them. Why did some of those patients get an interposition graft? Is that part of a recent change in strategy, or are you still recommending a patch as the primary procedure?
DR BROWN: Carl, most of our patients who had interposition grafts had aneurysms after their initial repair, and at redo I wanted to get rid of all of the abnormal aortic tissue. So, some of our patients who came back for recoartation did have aneurysms. And the majority of the aneurysms occurred after the first or initial surgery using a prosthetic patch.
DR BACKER: The other thing I wanted you to just mention was—because I've taken out some of these stents that the cardiologists have put in, and I have to say it's not a very fun day; so far the ones that I've done, I've ended up doing with partial bypass because of the risk of paraplegia now that they've got good flow—so maybe you could tell us, I think you had 3 patients that you took stents out of, maybe you could tell us a little bit about those 3 patients.
DR BROWN: One of our stent patients had migration and obviously we had to go after the stent and take it out. The other stent removals occurred late after initial intervention. One of our stent patients had a mycotic aneurysm around the stent. This child had his coarctation fixed at age 5 and had a recurrence, had a stent put in, and about a month later he became febrile and presented very ill, and the chest x-ray film showed a huge mass in the region of his distal aortic arch. We took him to the operating room, and there was a huge amount of pus surrounding that portion of his aorta. That entire portion of his aorta had to be resected and replaced with an aortic homograft. That patient was placed on left heart bypass for us to be able to get proximal and distal control. And the other patient was a patient who had an aneurysm after the stent was put in and probably before the days of stent grafts. After the aneurysm developed, the cardiologists felt this patient needed to go to the operating room and have the aneurysm resected.
But you raise an extremely important point, that when you go back in on a patient who has had angioplasty or stenting, or both, the collaterals have regressed and the risks for paraplegia are higher. We use a somatosensory evoked potentials on all of our coarctations and have for 20 years, and so we use that as a guide as to whether we need bypass or not.
DR FORBESS: I would add that, because it's so treacherous heading right back in the same pathway—just an advertisement for what we presented at the CHSS—we've gone to the right chest and just put a graft from the ascending aorta to the descending aorta and left the stent in place rather than try and extricate it.
DR BROWN: I recently had an opportunity to do an extra-anatomic bypass of a recurrent coarctation a couple of weeks ago, but that patient is not in this database.
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  B. Zipfel, P. Ewert, S. Buz, A. A. El Al, R. Hammerschmidt, and R. Hetzer Endovascular Stent-Graft Repair of Late Pseudoaneurysms After Surgery for Aortic Coarctation Ann. Thorac. Surg., January 1, 2011; 91(1): 85 - 91. [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 |
