| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Division of Cardiothoracic Surgery, University of Pennsylvania Medical Center, Philadelphia, Pennsylvania
b Cardiovascular Medicine, University of Pennsylvania Medical Center, Philadelphia, Pennsylvania
Accepted for publication July 9, 2007.
* Address correspondence to Dr Pochettino, Division of Cardiothoracic Surgery, Hospital of the University of Pennsylvania, 3400 Spruce St, Philadelphia, PA 19104-4283 (Email: alberto.pochettino{at}uphs.upenn.edu).
Presented at the Forty-third Annual Meeting of The Society of Thoracic Surgeons, San Diego, CA, Jan 29–31, 2007.
| Dr Bavaria discloses that he has a financial relationship with CarboMedics Inc, CryoLife Inc, Medtronic USA Inc, St. Jude Medical Inc, and Vascutek USA Inc.
|
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsCommentDiscussionAppendix 1Appendix 2AcknowledgmentsReferences |
|---|
Methods: From 1993 to 2004, 221 consecutive patients underwent repair of acute type A aortic dissection at our aortic center. Hemiarch repair was performed in 97.7% (216 of 221), and total arch in 2.3% (5 of 221). Of these, 72.9% (161 of 221) underwent aortic valve resuspension, and 27.1% (60 of 221) had aortic root replacement.
Results: In-hospital mortality for a primary operation was 12.7% (28 of 221). Actuarial survival was 79.2% at 1 year, 62.8% at 5 years, and 46.3% at 10 years. Significant risk factors for decreased survival included prior stroke, cerebral malperfusion, and length of cardiopulmonary bypass. Freedom from proximal reoperation after aortic valve resuspension was 94.6% at 5 years and 76.8% at 10 years, with cardiac malperfusion as the main risk factor. Freedom from distal reoperation was 87.6% at 5 years and 76.4% at 10 years, with Marfan syndrome, age, and extent of dissection as significant risk factors for reoperation. In-hospital mortality was 18.2% (2 of 11) after proximal reoperation and 31.2% (5 of 16) after distal reoperation.
Conclusions: We report improved long-term durability of our proximal root repair, with cardiac malperfusion as a significant risk factor. Marfan disease, younger age, and DeBakey type I dissection are risk factors for distal reoperation. To further improve long-term outcome, means to prevent progression of distal aortic disease need to be developed.
Acute type A dissection remains one of the most challenging diseases the cardiothoracic surgeon faces. In 1972, the estimated mortality without surgical treatment for acute aortic dissection was 1% to 2% per hour, with less than 10% surviving 3 days. Patients died from aortic rupture, severe aortic insufficiency, or malperfusion syndrome [1]. In a recent population-based series, the incidence of acute aortic dissection (both Stanford type A and B) was 3.5 per 100,000 person-years, with an overall 5-year survival rate of 32% and a median survival of 3 days [2]. In-hospital mortality after surgical treatment of acute type A aortic dissection is 14% to 32.5% in published series [3–6]. Without surgical treatment, the in-hospital mortality for acute type A aortic dissection is 58% [6]. Expeditious surgical treatment is therefore essential in successful management of patients with acute type A aortic dissection.
When the Thoracic Aortic Disease Center at the University of Pennsylvania Health System was established in 1993, our group initiated a uniform approach in the treatment of type A aortic dissection [7]. Our surgical protocol addresses replacement of the ascending aorta, repair or replacement of the aortic sinus segment to treat or prevent coronary malperfusion, resuspension or replacement of the aortic valve (AV), and replacement of most of the aortic arch to prevent malperfusion of the arch vessels and establish normal true lumen flow down the proximal descending thoracic aorta.
Open arch reconstruction is performed with uniform use of hypothermic circulatory arrest and retrograde cerebral perfusion. Antegrade cerebral perfusion is used selectively when circulatory arrest time is expected to be excessive, mainly for total arch replacement. We believe this approach provides optimal cerebral protection and allows accurate identification and resection of the primary tear site along with ease of arch reconstruction. Status of brain function and arch vessel perfusion are closely monitored by electroencephalography, when available, and bilateral carotid duplex ultrasound interrogation.
Short-term benefits of our standardized approach have been previously described [8]. In this report we wanted to define long-term outcomes of our strategy, focusing primarily on survival and need for proximal or distal reoperations. Prior reports of long-term outcomes include mostly heterogenous techniques and populations.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAppendix 1Appendix 2AcknowledgmentsReferences |
|---|
During the time period, 244 consecutive patients underwent repair for acute type A aortic dissection at the University of Pennsylvania, of whom 23 were excluded because HCA/RCP and open arch reconstruction were not used (n = 10) and/or a different proximal surgical technique was used (n = 18). These cases occurred in decreasing frequency during the first 5 years of the study because surgeons outside the Thoracic Aortic Program eventually stopped repairing acute type A dissections. The analysis included 221 cases.
Demographics and Clinical Presentation
The demographics and clinical presentation for the 221 patients are summarized in Table 1. Their mean age was 61.6 years (range, 21 to 89 years). Malperfusion syndromes were evident in 59 (26.7%), with myocardial ischemia noted in 7.2% of cases and cerebral malperfusion with evidence of preoperative neurological event in 7.2%. Preoperative aortic insufficiency (AI) evaluated by transesophageal echocardiogram (TEE) was graded 0 in 21.7% of patients, 1 to 2+ in 51.7%, and 3 to 4+ in 25.6%.
|
The mean ± SD cardiopulmonary bypass time was 238.8 ± 62.1 minutes, cross-clamp time was 169.3 ± 46.6 minutes, and hypothermic circulatory arrest with retrograde cerebral perfusion was 38.2 ± 14.5 minutes. Selective antegrade perfusion, used in the 3 patients with total arch replacement, averaged 43.3 ± 17.1 minutes.
Proximal repairs in the 221 patients included 161 (72.9%) with successful AV resuspension, with 52 (23.5%) requiring root replacement. One patient had valve-sparing root replacement; 4 underwent AV replacement within a nondissected or a repaired root; and 3 had prior well-functioning prosthetic AVs and therefore only had ascending graft replacement. Indications for root replacement are summarized in Table 1. Patients who underwent composite root replacement were more likely to have AI exceeding 2+ than those who underwent AV resuspension: 43.8% and 25.6%, respectively (p = 0.005). Six of 10 Marfan patients had composite root replacement and the other 4 underwent AV resuspension, with most of the diagnoses of Marfan syndrome made after repair (Ghent criteria).
Three of 22 patients with bicuspid AV underwent AV resuspension, and the other 19 had aortic root replacement. Hemiarch replacement was used in all but 3 patients who underwent total arch replacement.
Other associated procedures performed at the same setting included coronary artery bypass grafting (CABG) in 17 (7.7%), mostly for coronary malperfusion or when the right coronary artery could not be reimplanted within a root replacement. One patient (0.5%) had mitral valve repair, and 10 (4.5%) had ileofemoral revascularization.
We started using BioGlue surgical adhesive (CryoLife Inc, Kennesaw, GA) in February 1999, and for the entire series, it was used in 49.8% of cases. Of patients undergoing operation since February 1999, we used BioGlue in 67.5% of cases. We used a small amount of BioGlue in between the dissected layers to secure the Teflon felt neomedia as well as small amount externally for hemostasis.
Follow-Up
Mortality was assessed through medical records and the Social Security Death Index completed for October 30, 2006. Survival data were available for all but 1 patient. Survival follow-up was 904 patient-years, with mean follow-up of 49.1 months. The longest follow-up was 13.1 years. Clinical follow-up was obtained through medical records, telephone calls, and letters. Two patients were lost to clinical follow-up, with completed records in 219 of 221 patients (99%). Eleven patients had less than 1 year of follow-up. Total follow-up was 731 patient-years, with mean follow-up of 39.7 months.
Statistical Analysis
This was a retrospective observational study. Definitions are provided in Appendix 1. Variables are shown in Appendix 2 and were collected by retrospective review of patient hospital charts and office notes, by phone calls, and by using prospectively collected Society of Thoracic Surgeons (STS) data at the University of Pennsylvania. Continuous variables were expressed as the mean ± SD and were compared with an unpaired 2-tailed t test. Categoric variables, expressed as percentage, were analyzed with a 
2 test. Survival and freedom from reoperations were analyzed with the Kaplan-Meier actuarial method and compared with the log-rank test. Results at 1 year, 5 years, and 10 years are expressed as percentage ± SD. To analyze risk factors for long-term survival, proximal reoperations, and distal reoperations, univariate analysis was performed on patient demographics, presenting symptoms, intraoperative factors, and postoperative factors by comparing different subsets of patients. Significant or marginally significant (p 
0.20) risk factors were then analyzed by Cox multivariate proportional hazard regression and expressed as a hazard ratio (HR) with the 95% confidence interval (CI).
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAppendix 1Appendix 2AcknowledgmentsReferences |
|---|
Survival
The in-hospital mortality was 12.7%, and the 30-day mortality was 11.8%. The in-hospital mortality was 23.1% for root replacement and 8.1% for AV resuspension (p = 0.004). For the 28 patients who died in-hospital, the primary reason was cardiac failure in 7 (25%), exsanguination in 3 (10.7%), MSOF in 11 (39.8%), and neurologic events in 7 (25%). The actuarial survival for all patients was 79.2% ± 2.7% at 1 year, 62.8% ± 3.5% at 5 years, and 46.3% ± 5.1% at 10 years (Fig 1).
|
Univariate predictors of decreased long-term survival included peripheral vascular disease (p = 0.026), prior stroke (p = 0.024), coronary artery disease (p = 0.020), hypotension at presentation (p = 0.024), preoperative arrest (p = 0.003), cerebral malperfusion (p = 0.001), and length of CPB (p = 0.036). Multivariate analysis was performed on these additional, marginally significant variables: prior CABG, chronic renal failure, age, sex, and type of proximal reoperation. Multivariate proportional hazard regression analysis revealed that prior stroke, cerebral malperfusion, and length of CPB were significant risk factors for decreased long-term survival (Table 2).
|
Reoperations
Twenty-four reoperations were required: 8 patients required proximal reoperation only, 13 required distal reoperation only, and 3 needed both. One of the patients that required both underwent sequential operations: first, an AV replacement for endocarditis, and a year later, repair of 8.5 cm thoracoabdominal aortic aneurysm (TAAA). The other 2 patients had simultaneous operations through sternotomy: a redo ascending and hemiarch grafts for proximal and distal pseudoaneurysm, and aortic root replacement for severe AI and redo hemiarch for distal pseudoaneurysm. Freedom from any reoperation for all cases estimated by actuarial methods was 87.4% ± 3.2% at 5 years and 70.0% ± 7.4% at 10 years.
Proximal Reoperations
Indications for the 11 proximal reoperations (Table 3) included severe AI in 5 patients, pseudoaneurysm in 5, and endocarditis and graft infection in 3. Endocarditis developed in one of the separate AV replacement/valve grafts (Wheat) and in one patient with composite mechanical root replacement. No endocarditis developed in any of the patients who had AV resuspension, but one required reoperation for ascending graft infection.
|
|
Significant univariate risk factors for proximal reoperation in patients who had AV resuspension were cardiac malperfusion (p = 0.007) and ileofemoral malperfusion (p = 0.05). Freedom from proximal reoperations in patients without cardiac malperfusion was 95.7% ± 2.1% at 5 years and 79.7% ± 9.3% at 10 years. In patients with cardiac malperfusion, freedom from proximal reoperations was 75.0% ± 21.6% at 5 years and 37.5% ± 28.7% at 10 years (log-rank, 9.91; p = 0.0016; Fig 3).
|
|
|
|
|
|
Comment |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAppendix 1Appendix 2AcknowledgmentsReferences |
|---|
Our practice is immediate operative repair, regardless of the patient’s condition and the time of the day. We do not delay operative intervention with additional preoperative workup, including coronary and cerebral imaging. The patient is directly admitted to the operating room.
Our operative approach consists of replacing the entire ascending aorta, resuspension of the AV with repair or replacement of the sinus segment, and routine open replacement of the arch. This replacement is done under hypothermic circulatory arrest with retrograde cerebral perfusion. The false lumen is obliterated at the distal arch/proximal descending thoracic aorta, thus reestablishing normal flow in the descending thoracic true lumen. We use root replacement and total arch replacement selectively. Although this is a case series, the strength of the study is that the surgical techniques and the perfusion strategy are standardized, which results in a homogenous cohort.
Our hope was that our approach might result in improved freedom from proximal and distal reoperations and increased long-term survival. Our patients have characteristics similar to previous reports, including the International Registry of Acute Aortic Dissection database [6]. This study confirms that we substantially improved short-term survival, as measured by in-hospital mortality rate of 12.7%, compared with other reports. It is important to point out that all patients who arrived in the operating room with vital signs were included in our mortality analysis. Other series report an in-hospital mortality rate of 14% to 32.5% [3, 4, 6].
Complications after operative repair for acute type A dissection were significant. New postoperative strokes occurred in 7.4% of patients, which is comparable to stroke rates after elective aortic surgery of 8.1% to 8.7% at our institution as well as others [10, 11].
Long-term survival in our series was 79.2% at 1 year, 62.8% at 5 years, and 46.3% at 10 years. These finding are very similar to other reports, which range from 60% to 84% at 1 year [4, 12–16], 45% to 72% at 5 years [3, 4, 12–17], and 37% to 56% at 10 years [3, 4, 13, 15–17]. In the 4 patients who underwent separate AV replacement and ascending graft replacement with retention of the sinus segment, there was a trend toward worse long-term outcome both in survival and need for proximal reoperations. We believe that procedure should only be preformed in compromised patients in whom AV resuspension has failed yet sinus repair is successful.
On univariate analysis, long-term survival is primarily dependent on patient substrate and underlying diseases, agreeing with other reports [12]. Multivariate analysis found the strongest risk factors associated with decreased survival were perioperative neurologic status, including prior history of stroke and cerebral malperfusion. The length of CPB was also a significant but weaker risk factor. The use of composite valve graft has been associated with decreased early survival [3]. Our data confirm that in-hospital mortality is significantly higher for patient who underwent composite root replacement compared with AV resuspension. The reason remains unclear, even though requirement for root replacement may be a marker for more severe dissection and clearly lengthens the operative time. The long-term survival for these two groups, however, was not different.
The freedom from reoperations after AV resuspension has not been clearly established:
We put significant effort in obtaining a complete follow-up, and only 2 patients were lost to follow-up. Eleven patients (9 had AV resuspension) required proximal reoperation, most commonly due to severe AI or pseudoaneurysm at the proximal suture line. The freedom from proximal reoperation after all operations was 95.1% at 5 years and 77.8% at 10 years. The freedom from proximal reoperation in the patients that underwent AV resuspension was 94.6% at 5 years and 76.8% at 10 years.
Preoperative AI was present in 78% of patients. Worse AI (3 to 4+) resulted in more root replacements. However, preoperative AI grade exceeding 2+ in patients who underwent AV resuspension was not a predictor for proximal reoperation, contrary to other reports [4, 19], demonstrating that in properly selected patients with severe AI, AV resuspension results in acceptable long-term durability. Grade of AI after repair was never more than 2+ in our series. Furthermore, the grade of AI immediately after repair did not predict need for proximal reoperation.
Coronary malperfusion was associated with higher need for reoperation after AV resuspension on both univariate and multivariate analysis. The results of this study indicate that this patient population should undergo root replacement because coronary malperfusion appears to be a marker of severity of destruction of the root. The role of valve-sparing root replacement in type A dissection is an attractive option with short-term and mid-term outcomes comparable with AV resuspension techniques [20], but the long-term outcome is far from clear.
The fate of the aorta distal to the replaced segment after acute type A dissection repair has also not been defined in the current era:
In our series, 16 patients required distal reoperation, most commonly due to aneurysmal dilatation of residual descending aortic dissection but also for pseudoaneurysm at the distal suture line. The freedom from distal reoperation was 86.5% at 5 years and 75.4% at 10 years. Younger age (<45 years) and DeBakey type I dissection were significant risk factors for distal reoperation by multivariate analysis, and Marfan syndrome was a significant risk factor on univariate analysis.
It is difficult to compare these reports owing to potential differences in selection of patients that undergo replacement of thoracoabdominal aorta, a procedure generally associated with a high incidence of complications and often performed only as a last resort. We have followed up all of our residual dissections with serial imaging at least yearly and have offered surgery if yearly growth exceeds 0.75 cm or when the absolute maximal diameter exceeds 6.5 cm. Six of the distal reoperations, however, were performed to treat symptomatic distal suture line pseudoaneurysm or ruptures.
The high reoperation rate in this and other series is of concern, as is the high mortality rate of 31.2% after distal reoperations. Our results support a change toward more frequent use of total arch replacement with elephant trunk in patients younger than 45 and in patients with Marfan disease. Our database did not allow us to assess for the presence of persistent false lumen in the residual aorta or quantify size or growth rate of the descending aneurysms in our patients.
In addition to a greater use of total arch replacement in younger patients or those identified with collagen-vascular disease, we have started to address the residual dissection in the aorta distal to the replaced arch by deploying a covered stent graft into the distal arch/proximal descending thoracic aorta during open arch repair. The long-term outcome of stenting the residual dissected proximal descending aorta remains to be seen, and we plan to report our data as they become available. Our hypothesis is that acute stenting of the descending thoracic aorta in DeBakey I dissections will significantly decrease the need for distal reoperations.
In conclusion, we report improved in-hospital and long-term survival after our operative approach for acute type A aortic dissection. Significant risk factors for mortality include history of stroke, cerebral malperfusion, and length of CPB. Our standardized technique has improved long-term durability of the repaired residual proximal aorta and root, resulting in improved freedom from proximal reoperation.
Our data support that AV resuspension techniques are durable and have the same long-term survival as root replacement. The selection of the type of proximal operation during initial repair should depend on the condition of the valve leaflets and the extent of root involvement. Patients with coronary malperfusion have higher risk of root reoperation and should generally have root replacement.
Our distal reoperation rate is still significant and, furthermore, distal reoperations remain highly morbid procedures. Patients younger than 45, with DeBakey type I dissection and Marfan syndrome, have a higher risk of distal reoperation and therefore should be considered for total arch replacement. To further improve outcomes, means to prevent progression of distal aortic disease need to be developed, possibly consisting of stent graft deployment.
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAppendix 1Appendix 2AcknowledgmentsReferences |
|---|
DR POCHETTINO: We have a neurologist involved in all of our patients. Sixty percent of the patients were monitored intraoperatively. At present our neurologists can monitor the intraoperative EEG from home via a Web-based system. The same neurologists who perform the intraoperative monitoring follow all patients and make the clinical call whether a given patient has had a stroke. Any patient who has a clinical change detected by our neurologic team undergoes a CT scan. If there is no neurologic change, a CT scan is not necessarily performed.
Regarding long-term aortic surveillance, our intent is to follow all patients. Some patients do not come back to us. The vast majority do come back and are seen at 3, 6, and 12 months post-op with CT scans and echos. Thereafter, if their aorta is stable, we follow them once per year with CT scans and echos for life.
|
Appendix 1 |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAppendix 1Appendix 2AcknowledgmentsReferences |
|---|
|
Appendix 2 |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAppendix 1Appendix 2AcknowledgmentsReferences |
|---|
Type of proximal repair: aortic valve resuspension with ascending graft, aortic valve replacement with ascending graft, valve sparing root replacement with ascending graft, composite root replacement (mechanical or biologic) and ascending graft only;
Type of distal arch repair: hemiarch or total arch replacement.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAppendix 1Appendix 2AcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAppendix 1Appendix 2AcknowledgmentsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. A. Elefteriades Invited commentary. Ann. Thorac. Surg., September 1, 2009; 88(3): 779 - 780. [Full Text] [PDF]
|
|
|
|
|
|
A. Pochettino, W. T. Brinkman, P. Moeller, W. Y. Szeto, W. Moser, K. Cornelius, F. W. Bowen, Y. J. Woo, and J. E. Bavaria Antegrade thoracic stent grafting during repair of acute DeBakey I dissection prevents development of thoracoabdominal aortic aneurysms. Ann. Thorac. Surg., August 1, 2009; 88(2): 482 - 489. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Tanaka, K. Okada, Y. Kawanishi, M. Matsumori, and Y. Okita Clinical significance of anastomotic leak in ascending aortic replacement for acute aortic dissection Interactive CardioVascular and Thoracic Surgery, August 1, 2009; 9(2): 209 - 212. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Shimamura, T. Kuratani, G. Matsumiya, Y. Shirakawa, M. Takeuchi, H. Takano, and Y. Sawa Hybrid endovascular aortic arch repair using branched endoprosthesis: The second-generation "branched" open stent-grafting technique J. Thorac. Cardiovasc. Surg., July 1, 2009; 138(1): 46 - 53. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Okada, A. Tanaka, H. Munakata, M. Matsumori, Y. Morimoto, Y. Tanaka, T. Maehara, and Y. Okita Extended replacement of aortic arch aneurysms through left posterolateral thoracotomy Eur. J. Cardiothorac. Surg., February 1, 2009; 35(2): 270 - 275. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K.-H. Park, C. Lim, J. H. Choi, E. Chung, S. I. Choi, E. J. Chun, and K. Sung Midterm Change of Descending Aortic False Lumen After Repair of Acute Type I Dissection Ann. Thorac. Surg., January 1, 2009; 87(1): 103 - 108. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. T. Kouchoukos, P. Masetti, M. C. Mauney, M. C. Murphy, and C. F. Castner One-Stage Repair of Extensive Chronic Aortic Dissection Using the Arch-First Technique and Bilateral Anterior Thoracotomy Ann. Thorac. Surg., November 1, 2008; 86(5): 1502 - 1509. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. M. Sundt III, T. A. Orszulak, D. J. Cook, and H. V. Schaff Improving Results of Open Arch Replacement Ann. Thorac. Surg., September 1, 2008; 86(3): 787 - 796. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. Y. Szeto Invited Commentary Ann. Thorac. Surg., July 1, 2008; 86(1): 101 - 102. [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 |
