|
|
||||||||
J Thorac Cardiovasc Surg 2008;136:1289-1294
© 2008 The American Association for Thoracic Surgery
Evolving Technology |
a Department of Cardiac Surgery, Leipzig Heart Center, Leipzig, Germany
b Department of Angiology, Leipzig Heart Center, Leipzig, Germany
Received for publication January 20, 2008; revisions received April 7, 2008; accepted for publication May 19, 2008. * Address for reprints: Evaldas Girdauskas, MD, Leipzig Heart Center, Struempellstrasse 39, 04289 Leipzig, Germany. (Email: evagird{at}centras.lt).
| Abstract |
|---|
|
|
|---|
Methods: A total of 14 patients (8 male, mean age 59.8 ± 14.8 years) underwent conventional surgical therapy at our institution over a 5-year period after previous thoracic aortic stent implantation. The indications for surgery, intraoperative strategy, and perioperative and follow-up results were analyzed.
Results: The indication for stent implantation was type B aortic dissection in 10 patients, expanding degenerative thoracic aneurysm in 3 patients, and pseudoaneurysm in 1 patient. The median interval to conventional surgery after stent implantation was 4.5 months (range 0.1–49 months). The indication for surgery was persistent type I endoleak with progressive aneurysm enlargement in 7 patients, aortoesophageal fistula in 2 patients, retrograde type A dissection in 2 patients, infection of the endoprosthesis in 2 patients, and aortic valve insufficiency caused by perforation of noncoronary and right coronary cusps in 1 patient. The endograft had to be removed in 9 (64%) patients, and 5 (36%) patients required replacement of the thoracoabdominal aorta. In-hospital mortality was 7% (1 patient). No patients had a postoperative stroke or paraparesis. Eleven (79%) patients are alive after a mean follow-up of 13.5 ± 10.1 months (range 1–36 months).
Conclusions: Secondary surgical procedures after stenting of the thoracic aorta can be performed with very good results, despite the challenging clinical scenarios. Identification of successful surgical strategies for this difficult clinical problem is important in the era of increasing endovascular therapy.
| Introduction |
|---|
|
|
|---|
| Methods |
|---|
|
|
|---|
Endovascular Procedures
The indications for endovascular stent graft treatment of the thoracic aorta included type B aortic dissection in 10 patients (acute in 1 and chronic in 9), expanding degenerative thoracic aortic aneurysm in 3 patients (with contained rupture in 2 patients), and pseudoaneurysm in 1 patient. All patients with chronic type B aortic dissection had progressive dilatation of the perfused false channel in the descending thoracic aorta. The indication for endovascular therapy in acute type B dissection was subtotally narrowed true channel at the level of the lower abdominal aorta with signs of peripheral malperfusion. Four different stent graft systems were used over the study period: Talent (Medtronic Vascular, Santa Rosa, Calif) in 8 patients, Excluder (TAG; W. L. Gore & Associates, Inc, Flagstaff, Ariz) in 2 patients, Endofit (Endomed, Inc, Phoenix, Ariz) in 2 patients, and Valiant (Medtronic Vascular) in 2 patients. The proximal landing site was distal aortic arch (ie, proximal to the left subclavian artery) in 8 patients and descending thoracic aorta in 6 patients. Three patients required more than 1 stent graft implantation at the time of primary endovascular intervention. Secondary endovascular procedures were performed in 2 patients for persistent type I endoleaks 1 and 4 months after the primary intervention.
Indications for Surgery
The indications for secondary conventional surgical procedures included progressive enlargement of the aneurysm sac with a type I endoleak in 7 patients, retrograde type A aortic dissection in 2 patients, aortoesophageal fistula in 2 patients, infection of the endoprosthesis with sepsis in 2 patients, and significant aortic valve insufficiency after perforation of the noncoronary and right coronary cusps in 1 patient. The operation was performed on an emergency basis in 7 patients.
Surgical Procedures
The surgical technique used was dependent on the type of aortic disease (see below). Full cardiopulmonary bypass or isolated left heart bypass were used in all patients. The axillary artery was directly cannulated in most patients requiring surgery on the aortic arch. Selective antegrade cerebral perfusion was performed in most of these patients by clamping the brachiocephalic artery and by inserting a perfusion catheter into the left common carotid artery. The head was packed externally in ice during circulatory arrest.
In the 2 patients with retrograde type A dissection, diagnosed 1 week and 4 months after the endovascular procedure, a median sternotomy was performed. The aortic root procedure was performed during systemic cooling. When the systemic temperature reached 23°C, circulatory arrest was induced and antegrade selective cerebral perfusion was initiated. The distal aorta was trimmed to the level of the stent graft and an elephant trunk anastomosis, incorporating aortic wall and proximal edge of the stent graft, was accomplished. The supra-aortic orifices were implanted into the prosthesis as a Carrel patch.
In 2 patients with type Ia endoleak and isolated enlargement of the distal arch/proximal descending aorta, surgery was conducted in the same manner as described above. This approach via a median sternotomy was chosen only if the distal aorta was confirmed to have no endoleak and no enlargement on serial computed tomographic (CT) scans. For better exposure of the proximal descending aorta, the incision was extended superiorly in a left anterolateral direction. One patient had two previous endovascular reinterventions for type Ia endoleak with the proximal endograft extension abutting directly onto the bicarotid trunk (Figure 1 ). The uncovered end and proximal branches of the stent graft were removed during surgery to produce an adequate anastomotic line, which was performed with an elephant trunk technique as described above.
|
Open repair was required in 2 patients because of stent graft infection 3.5 months and 1.5 months after the endovascular procedure. The first patient had recurrent sepsis and a positive white blood cell scan. Because of the stent graft position in the mid–descending aorta extending to the diaphragm, a thoracoabdominal incision was used. Left heart bypass was performed with mild hypothermia (32°C) via cannulation of the femoral vessels. The whole descending aorta along with the infected endoprosthesis was removed and backbleeding intercostal arteries were oversewn. A silver-coated woven 22-mm Vascutek prosthesis (Sulzer Medica Ltd, Renfrewshire, Scotland) was used. The second patient had hemodynamic collapse necessitating cardiopulmonary resuscitation (CPR) 1 day before the operation, with documented migration of the stent graft and rupture of the distal aortic arch/proximal descending aorta ( Figure 2 ). Because of the proximal location of the endoprosthesis and the presence of concomitant coronary artery disease, a clamshell incision was performed. The distal arch and descending aorta were replaced with the aid of circulatory arrest at 24°C using an extra-anatomically (ie, retrocardially) positioned Dacron prosthesis. Inspection of the ruptured aneurysmal sac revealed a distally migrated stent with complete leakage of the posterior aortic wall. Both ends of the blind aneurysmal sac were oversewn and a saphenous vein bypass graft to the right coronary artery was performed during rewarming.
|
|
Statistical Analysis
Standard definitions were used for patient variables and outcomes. Operative mortality was defined as death during hospitalization. Categorical variables are expressed as percentages and continuous variables are expressed as mean ± SD throughout the manuscript. All statistical analyses were performed with the SPSS 14.0 software (SPSS, Inc, Chicago, Ill). Long-term survival was analyzed with the Kaplan–Meier method.
| Results |
|---|
|
|
|---|
The duration of cardiopulmonary bypass or left heart bypass was 162 ± 70 minutes. Aortic crossclamping was required in 10 patients with a mean time of 80 ± 38 minutes. Nine patients had to be operated on with the aid of circulatory arrest (mean duration 25 ± 17 minutes) at a rectal temperature of 22.3°C ± 1.1°C. Simultaneous selective antegrade cerebral perfusion was performed in 6 of these patients. The stent grafts were removed in 9 (64%) and left in situ in the remaining 5 patients.
One patient died while in the hospital, for a mortality rate of 7.1%. The predicted logistic EuroSCORE risk of mortality for the entire patient cohort was 39.4% ± 25.6%. The single patient who died was the one with aortoesophageal fistula who required reoperation for recurrent esophageal leakage and mediastinitis. He died of septic multiorgan failure after a 39-day stay in the intensive care unit.
The median length of stay in the intensive care unit was 3 days (range 1–39 days). Four (28.6%) patients required reoperation because of excessive postoperative bleeding. The median time supported by mechanical ventilation was 26 hours (range 7–960 hours). Two patients required temporary tracheostomy for prolonged weaning from the mechanical ventilator. No new neurologic deficits (stroke or paraplegia) were observed postoperatively. Preoperative paraplegia persisted in 1 patient who was admitted with a contained rupture of the distal false channel. The hospital stay ranged from 8 to 40 days postoperatively (mean stay 20.1 ± 10.1 days). Predischarge echocardiography in the 2 patients who underwent aortic valve repair (ie, pericardial patch repair and David procedure) revealed mild and trivial aortic insufficiency, respectively.
Follow-up information was available for all discharged patients. After a mean follow-up period of 13.5 ± 10.1 months (range 1–36 months), 11 patients were alive and doing well. No aortic reinterventions were required in any patient. A total of 2 patients died during follow-up. The first patient underwent repair of a type A dissection and died suddenly while in convalescence. No autopsy was performed. The second patient died 1.5 months after aortic arch replacement for type Ia endoleak as a result of necrotizing pancreatitis. One- and 2-year survival was 78% ± 11.3%.
| Discussion |
|---|
|
|
|---|
The reported incidence of retrograde type A dissection after thoracic stent graft implantation worldwide is approximately 1% to 2%, although some groups published a rate of 5% to 6.8%.3,10
Both of our patients with type A dissection underwent a prior endovascular procedure for chronic type B dissection with positioning of the proximal stent graft in the distal aortic arch.2,9,10
One of the 2 patients underwent endovascular reintervention for type Ia endoleak 1 month after initial stent graft deployment, with repeated balloon dilatations to achieve a tight seal. It is plausible that a type A dissection in this case was triggered by repeated endoaortic manipulations (eg, wire and sheath manipulation or balloon dilatation). Furthermore, an intramural hematoma involving the aortic arch and ascending aorta could be identified on the preprocedural CT scan retrospectively, performed before the initial endovascular procedure. Intramural hematomas in the stent graft landing areas in the distal aortic arch (ie, covering the left subclavian artery) in patients with type B dissection have been reported to predispose to formation of retrograde dissection.2,3
Excessive angulation (70°) of the aortic arch was also noted in our patient. The second patient reported having mild nonspecific chest pain after her endovascular procedure. We were surprised to find a retrograde type A dissection on routine CT scan performed 1 week after the intervention. This finding reinforces the importance of postprocedural surveillance CT scans.
Perforation of the right and noncoronary aortic valve cusps was seen in 1 of our patients. To the best of our knowledge, such a complication has not been previously published in the literature. The cause of this adverse event was presumably guidewire manipulation during stent implantation. Fortunately, valve repair with autologous pericardial patch augmentation was successful in this 23-year-old female patient.
Two patients in our series underwent surgery for endoprosthesis infection, which was confirmed by culture of the explanted endograft and aortic thrombus (Staphylococcus aureus in the first and Staphylococcus epidermidis in the second patient). The relatively short time interval after endovascular intervention (3.5 months and 1.5 months, respectively) indicates a periprocedural infection. Moreover, both of these patients had evidence of systemic inflammatory reaction after the endovascular intervention. An extensive inflammatory process of the posterior mediastinum with contained aortic rupture required an extra-anatomic solution in 1 patient. No signs of recurrent infection have occurred in these patients during follow-up. Alternatively, thoracic aortic homografts have been successfully used in selected cases of infected thoracic aortic fabric grafts, aortoesophageal fistulas, and other situations.11
Aortoesophageal fistula is an uncommon but well described complication after thoracic aortic stenting.4,5,12,13
Suggested mechanisms leading to aortoesophageal fistula formation are chronic endoleak leading to erosion into the adjacent esophagus, penetration of the stent graft through the aortic wall into the esophagus,4
and ischemic necrosis of the esophageal wall resulting from stent coverage of the arteries in the midesophageal segment.5
Our first patient had a stent deployed in the distal aortic arch because of descending aortic aneurysm and was monitored closely for 14 months because of secondary type Ib endoleak. Chronic endoleak was therefore the most likely mechanism of the aortoesophageal fistula. The second patient had a documented erosion of the esophageal mucosa before stent graft implantation, which had progressed to a large esophageal ulcer (4 x 3 cm) 3 months after endovascular sealing of a ruptured mid–descending thoracic aortic aneurysm. This clinical scenario is suggestive of ischemic pathogenesis of the fistula. Multiple combinations of treatment options have been used to deal with aortoesophageal fistula, including in situ arterial reconstruction, extra-anatomic bypass with concomitant primary esophageal repair, or esophagectomy with cervical esophagostomy and secondary restoration of gastrointestinal tract continuity. Conservative, nonsurgical therapy results invariably in a fatal outcome owing to massive hemorrhage or chronic mediastinitis.4
After having lost our first patient as a result of recurrent esophageal leakage 2 weeks after primary defect repair, we changed our surgical strategy toward a more radical approach consisting of esophageal resection and restoration of esophagogastric continuity using a gastric pull-up procedure at the second stage. Our second patient survived uneventfully.
Persistent type I endoleak with progressive increase in aneurysm size is one of the major limitations of endovascular treatment. In our series, all 7 patients operated on for type I endoleak underwent initial endovascular intervention for type B dissection (chronic dissection in 6 patients and acute in 1). Despite aggressive oversizing, primary type I endoleak was documented in 6 patients, although additional stent implantations and repeated balloon dilatations were tried to seal them during the initial procedure (Figure 4 ). Because of proximal stent extension into the distal aortic arch to cover the entry site close to the left subclavian artery, an extensive open repair with prolonged circulatory arrest was required to repair the aortic arch in these patients. However, it is important to note that none of the patients in our series had a perioperative stroke or new paraplegia. We believe our perioperative strategies, including selective antegrade cerebral perfusion, drainage of cerebrospinal fluid, and distal perfusion of the lower body enabled us to avoid neurologic complications in these complex aortic arch procedures.
|
Study Limitations
The limitation of our study is the small sample size, which markedly limits the statistical analyses that can be performed. However, to the best of our knowledge, our series is the largest to date in the literature. As the number of patients undergoing thoracic endovascular therapy continues to grow over time, we believe our in-depth description of the presentation and management of these patients will provide helpful insights for clinicians who are increasingly likely to encounter these very challenging clinical problems.
| Conclusions |
|---|
|
|
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
S. Leontyev, M. A. Borger, J.-F. Legare, D. Merk, J. Hahn, J. Seeburger, S. Lehmann, and F. W. Mohr Iatrogenic type A aortic dissection during cardiac procedures: early and late outcome in 48 patients Eur J Cardiothorac Surg, December 1, 2011; (2011) ezr070v1. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Albors, J. A. Bahamonde, J. M. Sanchis, R. Boix, and J. Palmero Aortoesophageal fistula after thoracic stent grafting Asian Cardiovasc Thorac Ann, October 1, 2011; 19(5): 352 - 356. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Dumfarth, M. Michel, J. Schmidli, G. Sodeck, M. Ehrlich, M. Grimm, T. Carrel, and M. Czerny Mechanisms of Failure and Outcome of Secondary Surgical Interventions After Thoracic Endovascular Aortic Repair (TEVAR) Ann. Thorac. Surg., April 1, 2011; 91(4): 1141 - 1146. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Garbade, M. Jenniches, M. A. Borger, M. J. Barten, D. Scheinert, M. Gutberlet, T. Walther, and F.-W. Mohr Outcome of patients suffering from acute type B aortic dissection: a retrospective single-centre analysis of 135 consecutive patients Eur J Cardiothorac Surg, September 1, 2010; 38(3): 285 - 292. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. Biasi, T. Ali, T. Loosemore, R. Morgan, I. Loftus, and M. Thompson Hybrid repair of complex thoracoabdominal aortic aneurysms using applied endovascular strategies combined with visceral and renal revascularization J. Thorac. Cardiovasc. Surg., December 1, 2009; 138(6): 1331 - 1338. [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 |