| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2004;78:837-845
© 2004 The Society of Thoracic Surgeons
a Department of Cardiothoracic and Vascular Surgery, The University of Texas at Houston Medical School, Memorial Hermann Hospital, Houston, Texas, USA
Accepted for publication March 30, 2004.
* Address reprint requests to Dr Safi, Department of Cardiothoracic and Vascular Surgery, The University of Texas at Houston Medical School, UTH Medical Center, 6410 Fannin St, Suite 450, Houston, TX 77030, USA.
Hazim.J.Safi{at}uth.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 methodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
METHODS: Between January 1991 and March 2003 we repaired aneurysms of the proximal aorta in 597 patients. Of these patients, 104 had reoperations for replacement of the ascending aorta, aortic root, or transverse aortic arch. Previous surgery was defined as any previous cardiac or proximal aortic repair. Median age was 60 years, and 29 of the patients (28%) were female. Indications for reoperation and replacement of the proximal aorta included acute type A dissection in 6 patients (5.8%), aneurysm with chronic dissection in 60 (57.7%), progression of aneurysm in 23 (22.1%), infection in 12 (1.5%), inflammatory disease in 2 (1.9%), and atheromatous disease in 1 (1.0%). Reoperations included aortic root replacement in 20 patients (19.2%), total arch replacement with elephant trunk in 28 (26.7%), ascending and proximal arch in 39 (37.5%), and ascending aorta in 27 (26.0%). The median interval between operations was 69 months. Retrograde cerebral perfusion was used in 80 (77%) cases.
RESULTS: Chronic dissection was the most common indicator for reoperation in our population, followed by progression of aneurysm and infection. Thirty-day and in-hospital mortality was 13.5% (14 of 104) and 15.4% (16 of 104), respectively. Chronic obstructive pulmonary disease, renal dysfunction, and increased pump time were risk factors for mortality. Median follow-up was 5.02 years. Eight patients died during that period. Estimated survival at 1, 5, and 10 years was 83%, 80%, and 62%, respectively. Freedom from second proximal reoperations was 97.1% (10 of 104). Freedom from subsequent distal thoracic aortic repair was 84.6% (8 of 104).
CONCLUSIONS: Reoperations of the ascending aorta and aortic arch can be performed safely with good long-term results. Patients with previous proximal aortic dissection repair need long-term surveillance. Renal dysfunction and chronic obstructive pulmonary disease must be carefully considered before reoperations of the proximal aorta.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
The purpose of this study was to examine our experience with reoperations for proximal thoracic aortic disease and to report the primary causes for reoperation, describe the procedures performed, and identify the determinants of early and late outcome.
|
Material and methods |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
|
All preoperative, intraoperative, and postoperative data were collected prospectively for a period of 13 years and entered into a database. Analysis was retrospective. Survival was ascertained by direct patient contact (telephone or letter) and by searching the social security death index.
Chronic obstructive pulmonary disease was defined by a history of chronic bronchitis and emphysema, or, while using bronchodilators, less than 60% of predicted forced expired volume in 1 second. A serum creatinine level of greater than 2.0 mg/dL or the need for dialysis defined renal dysfunction. Repairs involving dissection were considered acute if surgery was performed in less than 14 days from the onset of symptoms and chronic if after 14 days. Cerebrovascular disease was defined by a history of cerebrovascular accident or intervention for carotid artery disease. Actuarial survival was determined at 5 and 10 years for each of the factors analyzed. Disease-free actuarial survival (Table 1) is survival without further reoperations of the proximal aorta.
|
After sterile preparation, the previous skin incision is incised to the sternum, and the sternal wires are removed. The chest is entered inferiorly at the subxiphoid process. Up and outward retraction is applied to both sides of the xiphoid process, lifting the sternum off the mediastinal contents. Sharp dissection is then used to separate the mediastinal contents from the undersurface of the sternum. The left and right pleural spaces are entered. The process of meticulously cutting adhesions from the undersurface of the sternum followed by cutting the sternum for a distance of 5 to 6 cm using the saw (Fig 2A) continues in stages until complete separation of the sternum is achieved (Fig 2B). Further dissection of the anterior heart is performed if the left internal mammary artery was used for a previous coronary artery bypass graft procedure. After completion of the redo sternotomy the heart is prepared for cannulation.
|
After cannulation and the initiation of cardiopulmonary bypass, the patient is systemically cooled, monitoring both nasopharyngeal temperature (cerebral) and bladder temperature (core body). The superior vena cava is cannulated for RCP by means of the right atrium, and snares are applied to both the inferior and superior vena cavae. We depend on intensive neurologic monitoring using power M-mode transcranial Doppler to verify bilateral middle cerebral artery blood flow throughout the case [7]. This becomes particularly important in cases of dissection when during cardiopulmonary bypass uncertainty may exist as to whether perfusion is being successfully delivered to the cerebral arteries.
Circulatory arrest is used for aneurysms that originate in the mid-to-distal ascending aorta and for all dissections. This permits complete resection of the diseased aorta and an open distal anastomosis. A 10-lead electroencephalogram monitors cerebral function, and a near-infrared spectroscopy unit continuously monitors cerebral oxygenation. Once the electroencephalogram is isoelectric, which coincides with a nasopharyngeal temperature of 15° to 20°C, cardiopulmonary bypass is discontinued and circulation is arrested. Retrograde cerebral perfusion is begun through the superior vena cava cannula. In the past, we used a conventional flow rate of 500 mL/min in the RCP circuit and kept superior vena cava catheter pressure less than 25 mm Hg. Because we now use power M-mode transcranial Doppler to directly monitor cerebral blood flow in both middle cerebral arteries, this technique identifies any reversal of flow during RCP and provides a guide for optimal RCP flow.
For the ascending and arch portions of aneurysms that extend into the descending thoracic aorta, we perform the first stage of the elephant trunk technique. This technique, introduced by Borst and colleagues in 1982 [8], has undergone several modifications [9, 10], but has maintained its original objective, which is to circumvent cross-clamping the descending thoracic aorta during the second stage, avoiding dissection of the proximal descending thoracic aorta from the pulmonary artery and potential injury. In cases of severely atheromatous descending thoracic aortas we sometimes place a polyethylene terephthalate fiber (Dacron, Wichita, KS) tube graft into the proximal descending thoracic aorta and use another beveled graft to reconstruct the arch, rather than attaching the elephant trunk graft directly to the arch graft. This modified elephant trunk technique is also used for cases of chronic dissection after fenestration of the dissecting membrane has been performed. The modified elephant trunk prevents kinking and distortion of the transverse arch graft and allows insertion of separate sized grafts especially when dissection has left a small descending thoracic lumen (Fig 3). In cases of chronic dissection, fenestration of the dissecting membrane is important when performing the elephant trunk so as to ensure perfusion of both the true and false lumens (Fig 4).
|
|
In cases of previous coronary artery bypass grafting that used saphenous vein grafts, reattachment of the saphenous vein grafts may be accomplished in several different ways, depending on the length of the original saphenous vein graft. Reattachment options include a Carrell patch (button) to the aortic graft (Fig 5A), a modified Cabrol interposition graft as individual graft (Fig 5B) or as a group if multiple saphenous veins are in close proximity (Fig 5C), or using an interposition saphenous vein directly to the new ascending graft (Fig 5D). The modified Cabrol technique uses two interposition grafts to reattach each main coronary ostia to the ascending graft.
|
|
Results |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
Procedures performed included aortic root replacement in 20 patients (19.2%), total arch replacement with elephant trunk in 28 patients (26.7%), ascending and proximal arch in 39 patients (37.5%), and ascending aorta in 27 patients (26.0%). Of the aortic root replacements, we used the modified Cabrol most often (17 of 20) for reattachment of the coronary ostia. In addition, we performed 17 (16%) cases of coronary artery bypass grafting and 17 (16%) cases of aortic valve replacement. In 30 cases, we reattached the previously placed saphenous vein grafts to the newly placed ascending aortic graft. This was performed using the button technique in 50% (15 of 30) of cases, followed by interposition saphenous vein graft 27% (8 of 30) of cases and modified Cabrol in 23% (7 of 30) of cases. One patient had redo median sternotomy with replacement of the transverse aortic arch and a modified elephant trunk for symptomatic atheromatous disease (grade III and IV) of the aorta. Eleven patients (11%) had a history of aortic valve replacement for bicuspid aortic valve. Six patients (5.8%) had clinical features of Marfan's syndrome. Infections included either native or prosthetic endocarditis or aortic graft infection. Inflammatory disease included one case each of Takayasu's arteritis and Behçet's disease.
Postoperative complications included 7.7% (8 of 104) reoperation for bleeding, 18.3% (19 of 104) ventilatory failure requiring tracheostomy, 2.9% (3 of 104) permanent stroke, and 1.9% (2 of 104) transient ischemic attacks. There were no cases of paraplegia.
Mean patient follow-up was 3.3 years with a range of 3 to 128 months. Thirty-day and in-hospital mortality was 13.5% (14 of 104 patients) and 15.4% (16 patients), respectively. The median interval between the operations was 69 months with range of 1 to 256 months. Retrograde cerebral perfusion was used in 80 (77%) cases. The mean RCP time was 33.9 minutes (range, 3 to 89 minutes). Median follow-up was 5.02 years. Eight patients died during that period. Estimated survival at 1, 5, and 10 years was 83%, 80%, and 62%, respectively.
On univariate analysis, COPD, renal insufficiency, and cardiopulmonary bypass time were the only significant risk factors for mortality (Table 1). Table 1 reports disease-free survival, which is defined as survival without reoperation of the proximal aorta. Multivariable analysis identified COPD and renal dysfunction as the only independent predictors of early and late mortality (Table 2). Kaplan-Meier diagrams for disease-free survival are depicted in Figure 6.
|
|
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
The most significant factors associated with both decreased early and long-term survival were a history of COPD and renal dysfunction. In our recent report that examined all replacements of the ascending and arch, including reoperations, both factors significantly influenced both early and long-term survival [13]. In a review by Crawford and associates [12] of 717 patients, although a history of COPD was not significantly associated with early mortality, COPD was associated with diminished midterm survival (p < 0.0042). They similarly found renal dysfunction to be associated with diminished long-term survival.
The prospects of poor long-term survival in patients with renal dysfunction is well known [14]. It is not as clear, however, why COPD has such a significant detrimental affect on long-term survival during repairs of the proximal aorta. Direct injury to the pulmonary parenchyma during redo median sternotomy may play a role. Another consideration is the exposure of the already marginal lung parenchyma to a second episode of profound hypothermic circulatory arrest. Although much attention has been paid to cerebral and myocardial protection during profound hypothermic circulatory arrest, new focus needs to be placed on preserving lung function, especially in patients with marginal lung function. Otherwise, careful consideration of surgery in these patients should be assumed.
Aneurysm associated with chronic dissection after previous repair of acute type A aortic dissection followed by aneurysm progression after a cardiac procedure were the most common indicators for reoperation. Because other series have reported different predominant reasons for reoperation, this was most likely related to referral patterns as opposed to any other reason.
Forty-eight patients who presented for reoperation had had previous acute type A aortic dissection, 26 (54%) of which had been repaired by cross-clamping the dissected aorta without the use of profound hypothermic circulatory arrest. Although it would be difficult to prove that cross clamping played a significant contributory role in the need for reoperation, we advocate profound hypothermic circulatory arrest during these cases for characterization and for resection and repair of the tear. The fact that the majority of reoperations for ascending and transverse arch aorta were for chronic dissection underscores the importance of long-term serial follow-up for patients with acute and chronic dissection.
Cannulation technique varied depending on aortic disease. In reoperative proximal aneurysmal disease, we most often cannulate the femoral vessel for the establishment of cardiopulmonary bypass. However, in cases of severely atheromatous diseased proximal and descending thoracic aorta, we establish arterial cannulation by means of the subclavian artery or the ascending aneurysm directly if feasible. We use transesophageal echocardiography to evaluate the status of the aorta.
Cannulation of the axillary artery in cases of dissection can be problematic because of the potential for multisystem and cerebral malperfusion during cardiopulmonary bypass. We recently adopted the power M-mode transcranial Doppler ultrasound for intraoperative monitoring of bilateral middle cerebral arteries during all cases of proximal aorta reconstruction [7]. By using this device we are able to confirm bilateral cerebral perfusion during the entire procedure. This becomes especially important during cases of acute or chronic dissection when cerebral perfusion may become compromised at the initiation and during cardiopulmonary bypass as a result of abnormal flow patterns of dissected vessels. Any evidence of cerebral malperfusion will then lead us to alter our cannulation technique. Other clues to cerebral malperfusion include uneven cooling or warming as evidenced by irregular readings in the nasopharyngeal and core body temperature and also by unequal cerebral oximetry readings.
In cases of chronic dissection and extensive aortic aneurysm, the elephant trunk technique is used [8]. Oftentimes, identifying the true and false lumen can be difficult, but fenestration or resection of the dissecting membrane in the arch and proximal descending thoracic aorta is imperative if the graft in the descending thoracic aorta is to deliver blood flow to the true and false lumens. We often use a short elephant trunk (less than 10 cm) for the same reason. Of note, if the false lumen is thrombosed, we do not manipulate the lumen, but instead modify the procedure with the placement of a separate smaller graft into the descending thoracic aorta. Manipulation of a thrombosed false lumen can disrupt already tenuous intercostal artery and radicular artery circulation and result in paraplegia.
Once the distal reconstruction is completed, we reestablish antegrade flow during the rewarming phase with an arterial cannula in the arch graft. There is evidence that this antegrade flow may reduce neurologic injury during the rewarming phase [15]. The proximal reconstruction depends on previous surgery. In the majority of our cases, chronic dissection arising after repair of more proximal acute type A dissection was repaired using a short tube graft with resuspension of the aortic valve. The previous repair is inspected, paying close attention to the sinuses of Valsalva and the aortic valve. If normal without dysfunction, then the root is preserved, and either an end-to-end or end-to-side transverse arch graft to the previous ascending graft anastomosis is performed. As the great vessels tend to be fixed in the setting of reoperation, an end-to-side graft-to-graft anastomosis is sometimes preferred to prevent kinking.
Crawford and coworkers [12] advocated the removal of all aneurysmal diseased aorta in the initial surgery. Specifically, they recommended aneurysmal replacement if disease was identified at the time of cardiac surgery such as coronary artery bypass grafting or valve replacement or repair. They also recommended resection of any aneurysmal disease in cases of Marfan's syndrome. Based on these recommendations and our own experience we also maintain an aggressive approach to concurrent ascending aneurysmal disease in patients undergoing cardiac surgery. Currently, we will consider replacing the ascending aorta if an aortic diameter of 4.5 cm or greater is verified at the time of cardiac surgery.
In conclusion, aortic dissection most commonly initiated reoperation of the proximal aorta in our population. Progression of aneurysm and infection were the second and third most common causes. The only discernable risk factors for diminished survival were the presence of COPD and renal dysfunction. With careful chest entry and meticulous removal of adhesions, redo repair of the ascending aorta and arch can be performed safely with good long-term results.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
Now with the availability of our bioglues, it is my opinion that we have revolutionized the treatment of acute type A aortic dissection, and I think that we at least have become more aggressive in operating on these patients under circulatory arrest during their initial operation and dealing with the transverse arch, and it is my impression at this time that we may see fewer of these cases with late aneurysmal dilatation of the arch.
My questions to you are, number one, are you using these bioglues in the initial operation for type A dissection and have you seen improved results or therefore a decreased incidence of late aneurysms in these patients? Thank you.
DR ESTRERA: Thank you very much for your questions, Doctor Stowe. In general, we do have experience with bioglue, but we currently do not prefer to use it at the initial repair for acute type A dissection. What we do prefer for reconstruction of the ascending aorta in acute aortic dissection is the use of multiple interrupted pledgetted horizontal mattress sutures for reinforcement of both the proximal and distal anastomoses; our technique was presented during the video session at last year's meeting. We have performed approximately 170 acute type A dissections now, and our reoperation rate is approximately 2% for late proximal aneurysm formation. My concern with the bioglue in the long term—and we still are waiting for long-term results with the use of bioglue—is the formation of pseudoaneurysm and aneurysmal dilatation at the site where the bioglue has been applied to reattach the layers. We have reoperated on 3 patients in whom bioglue was used previously, and what is interesting is that there is not a lot of healing when you look at the aortic anastomoses. You actually remove the graft and you have a clump of bioglue and there is no adhesion to the native tissue, as opposed to when we use our pledgeted horizontal mattress suture technique, there is a lot of scarring there. So we have some reservations about the bioglue, and for these reasons we do not really use the bioglue. I have been satisfied with the technique that Doctor Safi taught me a few years ago.
DR FORREST RUBENSTEIN (New Orleans, LA): Tony, again, this is an excellent series of difficult patients that most of us are happy that you have operated on as opposed to us. I agree that using the open technique at the original operation and completely closing the false lumen without using a cross-clamp will obviate the need for some of these operations. I have two questions, one of which was along the line of Doctor Stowe's. My first question is, did any of these patients have their "failure" at the proximal end, either with the aortic valve or with the proximal dissection becoming aneurysmal? And number two, in regards to your fenestration, do you only do that in the patients in whom you are going to do an elephant trunk or in all of the patients?
DR ESTRERA: Thank you Forrest. To answer the second question first, usually most of the patients who develop chronic dissection with aneurysmal dilatation distal to the previous repair are often patients with extensive aortic aneurysm. So for that reason, we do fenestrate distally in most of these cases. If the descending thoracic aorta is not enlarged, there is really no need to fenestrate, so we try to avoid a lot of manipulation in the descending thoracic aorta. There have been two cases in which there was bleeding from the descending fenestration, and we have had to actually extend our incision across the left chest to repair that during the reoperation. With regards to your first question, we have seen a few cases of proximal (aortic root) aneurysmal dilatation. In our series of patients in whom we performed the original type A repair, we had to reoperate for progression of aneurysm at the root in two cases.
DR DENNIS M. MORITZ (Huntington, WV): In our own practice looking at these patients with disease of the ascending aorta, we are seeing an increasing number of patients in their late 70s, early 80s. Your oldest patient was 77. In evaluating patients for these operations, how do you account for age? I would have expected that you would have several octogenarians in a series this size. Thank you.
DR ESTRERA: That is a good question. This is our series of 104 patients; obviously the range was 16 to 77 years. We did observe that our mean age is actually younger than our mean age for all of our ascending arch repairs. The younger age may be related to our referral patterns. This younger age of presentation also underscores the importance of close follow-up in patients with acute dissection. The majority of these patients have a disease process involving the aortic wall and for this reason distal aneurysms may occur.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. E.W. Kirsch, T. Ooka, K. Zannis, J.-F. Deux, and D. Y. Loisance Bioprosthetic replacement of the ascending thoracic aorta: what are the options? Eur. J. Cardiothorac. Surg., January 1, 2009; 35(1): 77 - 82. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. L. Estrera, E. E. Porat, C. C. Miller III, R. Meada, P. E. Achouh, A. D. Irani, and H. J. Safi Outcomes of delayed sternal closure after complex aortic surgery Eur. J. Cardiothorac. Surg., June 1, 2008; 33(6): 1039 - 1042. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. L. Estrera, C. C. Miller III, M. A. Villa, T.-Y. Lee, R. Meada, A. Irani, A. Azizzadeh, S. Coogan, and H. J. Safi Proximal Reoperations After Repaired Acute Type A Aortic Dissection Ann. Thorac. Surg., May 1, 2007; 83(5): 1603 - 1609. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. N. Girardi, K. H. Krieger, C. A. Mack, L. Y. Lee, A. J. Tortolani, and O. W. Isom Reoperations on the ascending aorta and aortic root in patients with previous cardiac surgery. Ann. Thorac. Surg., October 1, 2006; 82(4): 1407 - 1412. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Jault, A. Rama, L. Lievre, N. Bonnet, P. Leprince, A. Pavie, and I. Gandjbakhch Chronic dissection of the ascending aorta: surgical results during a 20-year period (previous surgery excluded). Eur. J. Cardiothorac. Surg., June 1, 2006; 29(6): 1041 - 1045. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Aortic root replacement after previous surgical intervention on the aortic valve, aortic root, or ascending aorta. J. Thorac. Cardiovasc. Surg., March 1, 2006; 131(3): 601 - 608.
|
|
|
|
|
|
F. Dagenais, P. Voisine, and P. Mathieu Giant pseudoaneurysm after proximal aortic surgery treated by means of redo axillary artery cannulation and use of an Endoclamp device J. Thorac. Cardiovasc. Surg., July 1, 2005; 130(1): 208 - 209. [Full Text] [PDF]
|
|
|
|
|
|
R. De Paulis, E. Cetrano, M. Moscarelli, G. Ando, F. Bertoldo, R. Scaffa, F. Tomai, and L. Chiariello Effects of ascending aorta replacement on aortic root dilatation Eur. J. Cardiothorac. Surg., January 1, 2005; 27(1): 86 - 89. [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 |
