| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2004;78:926-932
© 2004 The Society of Thoracic Surgeons
a Division of Congenital Heart Surgery, Michael E. DeBakey Department of Surgery, Houston, Texas, USA
b Division of Pediatric Cardiology, Department of Pediatrics, Baylor College of Medicine, Houston, Texas, USA
Accepted for publication February 17, 2004.
* Address reprint requests to Dr DiBardino, Congenital Heart Surgery, Texas Children's Hospital, 6621 Fannin St, MC WT 19345H, Houston, TX 77030, USA
djd{at}bcm.tmc.edu
Presented at the Fiftieth Annual Meeting of the Southern Thoracic Surgical Association, Bonita Springs, FL, Nov 13–15, 2003.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
METHODS: Twenty-one patients presented with recurrent aortic arch obstruction. Mean age and weight were 7.8 ± 5.4 years (range, 0.21 to 15.2 years) and 30.6 ± 21.8 kg (range, 3.6 to 90 kg), respectively. Recurrence involved the aortic arch to some degree in each case, as the mean preoperative transverse aortic arch z score was –2.9 ± 1.6 (range, –7.0 to 0.1). Thoracotomy was possible in 2 patients, using re-resection with end-to-end anastomosis (n = 1) and patch aortoplasty (n = 1). The remaining 19 patients required median sternotomy, cardiopulmonary bypass, and deep hypothermic circulatory arrest for complete relief of obstruction by aortic arch advancement (n = 10), patch aortoplasty (n = 8), or interposition grafting (n = 1).
RESULTS: There was 1 hospital death. Invasive blood pressure monitoring revealed no residual arm-to-leg gradient in 19 patients and a 20-mm Hg gradient in 2 patients. There have been no late deaths. No patients have undergone subsequent aortic intervention, and all are asymptomatic up to 85 months postoperatively. Two patients are currently followed with a 10-mm Hg arm-to-leg blood pressure gradient.
CONCLUSIONS: Anatomic reconstruction for recurrent aortic obstruction can be safely accomplished in the majority of patients. We favor median sternotomy because of the ability of establishing cardiopulmonary bypass, the facility of anatomic reconstruction techniques, and the ability to repair concomitant cardiovascular lesions.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
Initial reports of surgery for recurrent aortic obstruction were discouraging (operative mortality 15% to 33%) [3, 4]. Although Castaneda and Norwood [5] reported mortality less than 10%, a residual gradient of 40 mm Hg or more was noted in 30% of patients immediately after reoperation. These early results prompted investigation into alternative repair strategies. Extraanatomic aortic bypass techniques, including lateroisthmic bypass grafting (subclavian artery or aortic arch to descending aorta) and ascending-to-descending aortic bypass, were initially met with enthusiasm because they avoided hazardous dissection, eliminated residual gradients, and produced good early results [6, 7]. Vijayanagar and associates [8] first described the technique of incising the posterior pericardium to gain access to the descending aorta for ascending-to-descending aortic bypass through a median sternotomy. This technique continues to be effective in selected adult patients, as recently reported by Connolly and associates [9] and Attenhofer-Jost and colleagues [10], and allows concomitant repair of intracardiac defects.
On the basis of our experience with primary anatomic arch repair in neonates and infants, we have used an alternative strategy for recurrent aortic obstruction. Our technique relies on anatomic reconstruction, typically through a median sternotomy and using cardiopulmonary bypass with deep hypothermic circulatory arrest (DHCA). We believe that this strategy allows complete relief of obstruction at the level of the aortic arch and safe, single-stage repair of concomitant intracardiac defects while avoiding or limiting conduit use in growing children. We describe the pattern of recurrence, our operative technique, and results.
|
Material and methods |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
Mean age at reoperation was 7.8 ± 5.4 years (range, 0.21 to 15.2 years), and mean weight was 30.6 ± 21.8 kg (range, 3.6 to 90 kg). Details concerning the initial aortic repair techniques are outlined in Table 1. Median age at primary repair was 9 days (range, 0.1 to 152.2 months), and median interval from primary repair to reoperation for recurrent aortic obstruction was 4.6 years (range, 0.18 to 14.2 years). At the time of referral to surgery, significant arm-to-leg blood pressure gradients were present in all but 1 patient in whom midarch obstruction and an aberrant right subclavian artery eliminated the possibility of a gradient (Table 1). The mean arm-to-leg gradient in the remaining patients was 35.5 ± 15.2 mm Hg (range, 15 to 70 mm Hg).
|
Preoperative cardiac catheterization was performed in 13 patients (62%) for hemodynamic assessment of associated lesions or further delineation of arch anatomy (Table 1). The mean gradient from ascending to descending aorta measured at the time of catheterization was 35.0 ± 21.0 mm Hg (range, 10 to 63 mm Hg). One patient was found to have a resting gradient of 10 mm Hg and underwent isoproterenol infusion with a resulting gradient increase to 46 mm Hg. The severity of arch obstruction was underestimated by the gradient in several cases secondary to significant obstruction at the subaortic and aortic valvar level. One patient, for example, was found to have a 54-mm Hg gradient from the left ventricle to the descending aorta, with a 10-mm Hg component in the aortic arch. The appearance of the arch suggested that the severe left ventricular outflow tract obstruction minimized the significance of the arch obstruction and both were, thus, addressed at surgery. Seven patients (33%) underwent balloon angioplasty procedures before surgical referral, and these procedures were unsuccessful in reducing the gradient in each case. Since 1998, we have increasingly used preoperative magnetic resonance angiography (MRA) for further delineation of the aortic arch and descending aorta. Nine patients (43%) underwent preoperative MRA, and Figure 1A demonstrates the typical appearance of recurrence in our series, with significant obstruction visible at the level of the midarch.
|
|
|
Results |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
Hospital course
There was 1 hospital mortality in a 15-year-old boy who experienced acute suture line dehiscence 2 days after aortic arch advancement. The mean time to extubation was 1.4 ± 1.1 days (range, 0 to 5 days). Eight patients required transient inotropic support, and mean time to weaning of all inotropic agents was 2.4 ± 2.0 days (range, 0 to 6 days) for these patients. Average intensive care unit stay was 3.1 ± 1.4 days (range, 2 to 7 days), and average time to hospital discharge was 6.1 ± 2.3 days (range, 3 to 12 days). Nitroprusside was used for intravenous blood pressure control in the immediate postoperative period with transition to angiotensin-converting enzyme inhibitor therapy once oral intake was established. Complications included the development of a new left vocal cord paralysis in a patient undergoing median sternotomy (n = 1), reintubation for presumed pneumonia (n = 1), and narcotic withdrawal requiring readmission to the intensive care unit (n = 1). Transient arrhythmia occurred in 2 patients, including an episode of nonsustained ventricular tachycardia resolving without specific therapy. No gross neurologic deficits were found at discharge.
Follow-up
There have been no late deaths. All patients remain in a program of close surveillance consisting of clinical and two-dimensional echocardiographic follow-up. The child undergoing cardiac transplantation experienced a severe episode of rejection requiring 3 days of mechanical circulatory support 10 months after patch aortoplasty. She made a full recovery and is currently without symptoms. The patient requiring pulmonary band replacement underwent successful band takedown 13 months after patch aortoplasty, and no residual ventricular septal defects were found at that time. This was the only patient to undergo follow-up left-heart cardiac catheterization, and direct blood pressure measurement of the ascending aorta and descending aorta revealed no gradient. A 5-year-old patient undergoing patch aortoplasty was diagnosed with epilepsy 1 year after operation and is currently well controlled on carbamazepine.
No patients have undergone subsequent aortic intervention, and all are asymptomatic from a cardiovascular standpoint at a mean follow-up of 26.2 ± 24.4 months (range, 0.33 to 85.2 months) postoperatively. Noninvasive arm and leg blood pressure measurements are obtained at clinic visits in all patients. One patient with no gradient immediately after surgery currently has a 10-mm Hg arm-to-leg gradient. Of the 2 patients with a 20-mm Hg gradient immediately after repair, one currently has a 10-mm Hg arm-to-leg gradient and underwent MRA that revealed no significant obstruction (Fig 1B), and the other currently has no arm-to-leg gradient.
At least one follow-up two-dimensional echocardiogram with Doppler flow interrogation has been obtained in 17 of 20 surviving patients whereas the last 3 patients still await follow-up imaging. No evidence of significant aortic obstruction is present in any patient. Both patients undergoing thoracotomy at reoperation currently demonstrate a peak velocity of 2.5 m/s across the aortic arch and isthmus. Of those undergoing median sternotomy, 1 patient has a peak velocity of 2.6 m/s and 1 patient a peak velocity of 2.3 m/s across the arch and isthmus after patch aortoplasty, whereas 1 patient has a peak velocity of 2.5 m/s after aortic arch advancement. Two patients have evidence of worsening combined subaortic and aortic valvar stenosis with acceleration of flow across the left ventricular outflow tract to 2 to 4 m/s and will likely require operative intervention in the future.
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
In our series, recurrent aortic obstruction was seen to involve the arch (to some degree) in every case, as evidenced be a mean preoperative aortic arch diameter nearly 3 standard deviations below expected. This was confirmed by catheterization, MRA studies, and intraoperative inspection. Additionally, the median age of initial repair was only 9 days old, and the median interval to recurrence was less than 5 years, when somatic growth was still occurring at a high rate. One possible interpretation of these data is that the recurrent aortic obstruction was related to impaired arch growth that manifested with increasing somatic growth. Another possible hypothesis is that inadequate relief of arch obstruction at the initial operation was the central factor in recurrence, also manifesting with increasing somatic growth. Although we only performed one of the primary operations in this series and so cannot be certain about the initial arch morphology, we believe that it is essential to completely address significant arch hypoplasia at the primary repair of neonatal coarctation. It is this belief that has led to our current strategy of using arch advancement for repair of neonatal coarctation with significant arch hypoplasia [1], and as experience accumulates, we hope to demonstrate the long-term benefits of this strategy.
Surveillance and imaging
Kappetein and associates [15] warn that the incidence of reoperation for recurrent aortic obstruction is a gross underestimate of the true incidence of recurrence. Despite a 5.8% reoperation rate, they found recurrence in 41% of patients surviving primary coarctation repair with blood pressure gradients of greater than 20 mm Hg in those undergoing catheterization and upper-to-lower limb gradients of more than 50 mm Hg in those undergoing exercise testing. This raises important questions concerning the optimal surveillance modality after repair. Therrien and associates [16] recently assessed a variety of clinical and imaging modalities and concluded that combined clinical assessment and MRA or clinical assessment and transthoracic two-dimensional echocardiogram with MRA reserved for suspicious findings were acceptable screening alternatives. We rely on the latter strategy and stress the need for lifelong cardiovascular surveillance after repair of aortic obstruction.
Options for recurrent arch obstruction
Balloon angioplasty has been increasingly used for recurrent aortic coarctation and has more recently been recommended as the procedure of choice by several centers [17]. The safety of this intervention was reinforced by the Valvuloplasty and Angioplasty of Congenital Anomalies Registry Data, which reported 4 deaths in 548 patients (0.7%) undergoing balloon dilation therapy for recurrent aortic obstruction [18]. Although it appears that isolated, short segment recurrent obstruction is amenable to balloon dilation, one third of the patients in our series had failed catheter intervention and were unlikely to have received long-term benefit from balloon dilation alone secondary to aortic arch hypoplasia and long segments of obstruction. Although stent placement can improve long-term patency after balloon dilation, we believe that transverse aortic arch stenting should be avoided in growing children.
When faced with recurrence we approach each patient with the intention of providing an anatomic repair. Anatomic reconstruction techniques for recurrent aortic obstruction have traditionally included resection and reanastomosis, patch aortoplasty, and interposition grafting [19–24]. Extraanatomic bypass techniques were born in an era of disappointing results for the operative treatment of recurrent coarctation. Although Connolly and associates [9] have recently obtained excellent results with ascending-to-descending aortic bypass through a median sternotomy in older patients (15 to 66 years of age), they warn against the use of this technique in children and cite the known risk of anastomotic dehiscence with somatic growth. Although bypass conduits have been used in patients 4 to 11 years of age at other centers [25–27], we agree with Connolly and associates that pediatric patients are not best served by extraanatomic bypass and recommend anatomic repair whenever possible.
The concept of aortic coarctation repair through a median sternotomy is not new, particularly in the face of recurrence or concomitant intracardiac defects [21–22, 28]. Because many patients with recurrent aortic obstruction have arch hypoplasia and concomitant intracardiac defects requiring repair, thoracotomy may not be the optimal approach. Two existing reports of patients with recurrent aortic obstruction have specifically described a subset of patients requiring attention to the aortic arch through a median sternotomy [21, 22]. Techniques included patch aortoplasty extended into the proximal arch and interposition grafting performed under DHCA, allowing complete relief of obstruction with excellent results. Our strategy deferrers from these in that we rely heavily on aortic arch advancement with the intention of providing complete relief of obstruction, minimizing potential recurrence at the level of arch, and avoiding conduit use. Although Younoszai and associates [29] and Lacour-Gayet and associates [30] describe similar operations performed by means of a thoracotomy and with a cross-clamp occluding the arch and left brachiocephalic vessels, we emphasize their descriptions that the descending aorta is anastomosed to the underside of the aortic arch and not to the ascending aorta, as in our technique. With each of these procedures, decreased elasticity in the aorta and scar formation may not allow sufficient mobilization of the descending aorta for a tension-free advancement in the older child or adult. We find patch aortoplasty to be an acceptable anatomic technique whenever the possibility of a tension-free anastomosis is in question.
Regarding the use of RLFP, this is a relatively new technique for our unit that we have been using regularly since late 2000 for neonatal aortic arch reconstruction. As experience accumulates, we have expanded our indications for this technique and have been encouraged by studies showing its feasibility in the pediatric population [31, 32]. Although the patients undergoing RLFP in this series and in our Norwood population (unpublished data) have enjoyed greatly diminished circulatory arrest times in comparison with those with conventional DHCA, there has not been any published, objective scientific evidence that the use of RLFP improves clinical outcomes. We remain encouraged by its use and report good gross neurologic outcome [11] but cannot speculate on the ability of RLFP to impact on the subtle long-term derangements known to occur with extended DHCA times.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
DR DIBARDINO: Thank you, Doctor Robicsek, for your comments. That is an excellent question, and it really drives home our central point. The Mayo Clinic has recently described a similar technique, an ascending to descending bypass conduit, which is performed through the posterior pericardium via median sternotomy. They can do it off bypass or use bypass for any concomitant intracardiac repair. It is a great technique with excellent results in their hands. Being that we are a congenital heart surgery service, most of the patients that we have seen are younger children. The average age was only 7.8 years and the youngest were only infants. So we believe in approaching younger patients with the intention of performing an anatomic or a viable repair that can provide for viable growth. We feel that this is important in the long-term management of these patients, many of whom may require several reoperations in their lifetime.
|
References |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. M Maddali, J. Valliattu, T. al Delamie, and S. Zacharias Selection of Monitoring Site and Outcome after Neonatal Coarctation Repair Asian Cardiovasc Thorac Ann, June 1, 2008; 16(3): 236 - 239. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. W. Rossano, E.O.B. Smith, C. D. Fraser Jr, E. D. McKenzie, A. C. Chang, A. Hemingway, J. F. Price, H. A. Dickerson, and A. R. Mott Adults Undergoing Cardiac Surgery at a Children's Hospital: An Analysis of Perioperative Morbidity Ann. Thorac. Surg., February 1, 2007; 83(2): 606 - 612. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Rosenthal Coarctation of the aorta from fetus to adult: curable condition or life long disease process? Heart, November 1, 2005; 91(11): 1495 - 1502. [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 |
