|
|
||||||||
Ann Thorac Surg 2004;78:2099-2105
© 2004 The Society of Thoracic Surgeons
a Center for Aortic Surgery and Marfan and Connective Tissue Disorder Clinic and Department of Thoracic and Cardiovascular Surgery
b Department of Biostatistics and Epidemiology, The Cleveland Clinic Foundation, Cleveland, Ohio, USA
Accepted for publication May 14, 2004.
* Address reprint requests to Dr Svensson, The Cleveland Clinic Foundation, Department of Thoracic and Cardiovascular Surgery, 9500 Euclid Ave/Desk F25, Cleveland, OH 44195 (E-mail: svenssl{at}ccf.org).
| Abstract |
|---|
|
|
|---|
METHODS: From January 1, 1990, to January 1, 2002, 108 patients with prior cardiovascular surgery (isolated coronary artery bypass grafting, 51%; isolated valve surgery, 21%; aortic aneurysm repair, 24%; and combinations of these in the remainder) underwent reoperation for aortic dissection (emergency operation for acute dissection in 24%). Mean age was 63 ± 13 years, and 85% were men. The interval since prior surgery ranged from 10 days to 22 years (median, 3.8 years). This was the third operation for 8%. Ascending aortic repair with or without aortic arch or descending aortic repair was performed in 40%, aortic valve replacement (n = 15) or repair (n = 17) with ascending aortic repair in 30%, aortic root replacement with or without aortic arch or descending aortic repair in 30%, and aortic arch with or without descending aortic repair in 1%. Circulatory arrest was used in 78%, with retrograde brain perfusion in 58%.
RESULTS: Hospital mortality was 6%, stroke 4%, renal failure 2%, and respiratory failure 7%. Survival at 30 days and 1, 3, 5, and 7 years was 93%, 85%, 74%, 63%, and 53%, respectively. Aortic reoperation was performed in 7 patients, with freedom from this event at 30 days and 1, 3, 5, and 7 years of 98%, 95%, 93%, 91%, and 89%., respectively
CONCLUSIONS: Aortic dissection after cardiovascular surgery is rare and can be managed with acceptable operative risks and good long-term survival. Need for subsequent aortic reoperation is uncommon.
| Introduction |
|---|
|
|
|---|
| Patients and Methods |
|---|
|
|
|---|
Follow-up information was collected by telephone interview or responses to mailed questionnaires. The common closing date for cross-sectional follow-up was December 2002. Follow-up was 100% complete, with a mean follow-up of 4.1 ± 2.9 years.
The study included 92 men (85%) and 16 women (15%), with a mean age of 63 ± 13 years at the time of aortic dissection repair; 6 (5.6%) had Marfan syndrome. Eight patients (8.3%) had more than one previous cardiovascular operation (two coronary artery bypass grafting [CABG] operations in 5, two mitral valve operations in 2, and combined CABG and ascending aorta replacement in 1). The previous operations were varied (Table 1). Operations for previous aortic dissection occurred in 20 patients (19%), including DeBakey type 1 or 2 in 17 patients and DeBakey type 3 in 3 patients. In addition, previous operation for true aneurysms was performed in 5 patients (4.6%), in the ascending aorta in 4 and in the aortic arch in 1. Other preoperative details are shown in Table 1.
|
|
Operation
All patients underwent redo median sternotomy except one who had a clamshell incision. No cardiac or other structural injuries were documented during chest re-entry. Free rupture of aortic dissection was observed in 3 (2.8%): 1 into the pericardial space and 2 into the pleural space. An additional 2 patients were hemodynamically unstable at reoperation.
Cannulation site was the femoral artery in 46 (43%), axillary or subclavian artery in 48 (44%), aorta or innominate artery in 6 (5.6%), and undocumented in 8 (7.4%). Deep hypothermic circulatory arrest was used in 84 patients (78%), retrograde brain perfusion in 63 (58%), and antegrade brain perfusion in 5 (4.6%). Details of the operations are shown in Table 2. Ascending aortic repair with or without arch (n = 9) or descending (n = 3) aortic repair was performed in 43 (40%), aortic valve repair (n = 17) or replacement (n = 15) with ascending aortic repair with or without arch repair in 32 (30%), aortic root replacement (composite graft or valve-sparing) with or without aortic arch (n = 2) repair in 32 (30%), and arch with descending aortic repair in 1 (0.9%).
|
Previously repaired aortic valves (6 patients) were all replaced at the time of aortic dissection repair. Among 16 patients who had previously undergone subcoronary aortic valve replacement, 9 (56.3%) received a new aortic valve, including 5 who underwent composite grafting. None of 5 patients who had previous aortic root replacement required repeat aortic valve surgery at the time of aortic dissection repair.
Because of the retrospective nature of this study and difficulty of obtaining original operative reports from outside hospitals, we could not determine the influence of factors such as proximal CABG attachment methods, cannulation techniques, aortic size, presence of tricuspid aortic valves, and aortic dissection repair methods on risk of developing aortic dissection.
Aortic Operative Strategy
Most patients with acute dissection were transferred from outlying hospitals having undergone computed tomography, magnetic resonance imaging, or echocardiograms for chest pain. Coronary angiography was obtained if not already performed. Patients were then usually scheduled for surgery the next day unless a leak was present. Elective repairs were usually performed for patients presenting with asymptomatic incidentally found dissections after previous operation, evidence of growth of the dissected arch, size larger than 5.5 cm of the remaining aorta, or symptoms such as hoarseness, dysphagia, and stridor. Currently, most patients undergo operations with subclavian artery inflow with an attached side graft, transfemoral vein or direct right atrial venous drainage, and deep hypothermic circulatory arrest. Antegrade or retrograde brain perfusion was used on a selective basis [1, 2]. Most chronic dissection patients now undergo elephant trunk two-stage procedures or extended left thoracoabdominal incisions [3]. Biologic glue and aprotinin are usually not used.
Data Analysis
Descriptive statistics included the mean and standard deviation for continuous variables when the distribution was symmetrical; median and 25th and 75th percentiles when it was skewed; and frequencies and percentages for categorical variables (Appendix). Nonparametric estimates of time-related events were obtained using the Kaplan-Meier method. A parametric method was used to resolve the number of phases of instantaneous risk for each time-related end point (hazard function) and to estimate shaping variables [4]. To identify risk factors for all-cause mortality at any time after operation, multivariable analysis was performed in the hazard function domain using bootstrap bagging as the variable selection method, with p of 0.05 for retention of variables in the models [5, 6], following principles described previously [7].
Confidence limits of proportions and time-related depictions are accompanied by confidence limits equivalent to 1 standard error (68%).
| Results |
|---|
|
|
|---|
Postoperative Complications
The most common postoperative complications were bleeding and infection (Table 3). Strokes occurred in 4 patients (3.7%; confidence limit, 1.9% to 6.7%), 1 of whom died. Of the 4 patients who experienced strokes, deep hypothermic circulatory arrest with retrograde brain perfusion was used in 3 and was not reported in 1. Among hospital survivors, median intensive care unit and postoperative hospital stays were 3 and 13 days, respectively.
|
Survival at 30 days and 1, 3, 5, and 7 years was 93%, 85%, 74%, 63%, and 53%, respectively (Fig 2). Risk factors for death in the late hazard phase were higher preoperative blood urea nitrogen level (p = 0.002) and longer circulatory arrest time (p = 0.0001), as shown in Figure 3.
|
|
Freedom from the aortic reoperation at 30 days and 1, 3, 5, and 7 years was 98%, 95%, 93%, 91%, and 89%, respectively (Fig 4). Five of the 7 aortic reoperations (71%) were performed within the first year of surgery.
|
| Comment |
|---|
|
|
|---|
These authors note that reoperation after primary aortic dissection repair has increased in recent years [10], a trend mirrored in our series. However, because our study was limited to patients who underwent operation, this likely reflects referral bias and not an increase in the incidence of aortic dissection after cardiovascular surgery. Some patients may have refused referral for surgery, some may have died before referral to surgery, and some may have died before reaching the hospital. Thus, the true incidence of aortic dissection after cardiovascular surgery is unknown. We believe that the increase in reoperations could be related to increased referrals because of the relatively low postoperative mortality and morbidity; our hospital mortality (5.6%) is similar to that reported for primary aortic dissection repair [9]. We speculate that this low mortality is related to improved surgical techniques, circulatory arrest methods, cardiopulmonary bypass technologies, adjunctive brain perfusion methods, and axillary or subclavian cannulation strategies [13].
Aortic dissection after aortic valve operation was studied by von Kodolitsch and colleagues [11]. Late dissection occurred in 0.6%. Prevalence of previous aortic valve surgery was 9% in patients with aortic dissection. The authors found that fragility of the aortic wall, aortic regurgitation, and aortic wall thinning were risk factors for aortic dissection. However, in our study characteristics of the aortic wall at previous surgery were difficult to obtain owing to lack of documentation, or because most of the previous operations were performed at outside hospitals. In another study of aortic dissection after aortic valve replacement, dissection occurred more frequently in patients whose ascending aorta exceeded 50 mm in diameter at the time of valve replacement [12]. The authors recommended prophylactic ascending aorta replacement because of the risks of future dissection. McDonald and colleagues [13] found that the risk factors for late rupture (n = 15) after aortic valve procedures in 109 patients were older age and greater ascending aortic size indexed to body surface area. They also concluded that an enlarged ascending aorta required repair.
There are few studies of aortic dissection after previous cardiovascular operations [14, 15]. Gillinov and colleagues [14], in an earlier study from our institution, found 56 ascending aortic dissections after previous cardiovascular surgery, with a hospital mortality of 14%. They concluded that rupture of an aortic dissection after such surgery was infrequent because of adhesions. In our study, only 3 free ruptures occurred, but all the patients died. Once free rupture occurs before reoperation, salvage is less likely, because with emergency surgery, chest re-entry, cannulation, and exposure of the aorta are more complicated and time consuming than in primary operation for aortic dissection. Our study showed a higher mortality with emergency surgery, but this could be caused by chance.
Previous operation in our study was predominantly for coronary artery disease. Because hemodynamic instability or free rupture in redo aortic dissection patients is rare, patients should be considered for preoperative coronary angiography to evaluate native coronary artery disease and prior CABG surgery. Our study found that catheterization did not increase hospital mortality. Patients who have had previous CABG are more likely to require concomitant bypass surgery (87%; p < 0.0001) at the time of aortic dissection, mainly because of progression of the coronary artery disease, graft occlusion, or technical difficulties with reimplantation of previous grafts. Indeed, 27% without known coronary artery disease required CABG procedures.
During an average of 4.1 years of follow-up, there were 7 aortic reoperations. Among these, 3 were for endocarditis (2 infections occurred after composite graft placement and another after aortic valve repair), and they were performed in the first postoperative year. Reoperation for a new aortic dissection was observed in only 1 patient, and it occurred 8 years after surgery. Thus, new aortic dissection after redo dissection surgery is rare; furthermore, the most common cause of late deaths was not related to aortic dissection. The association of renal dysfunction and lengthy circulatory arrest with long-term mortality is not surprising.
In this study, stroke occurred in 3.7% of patients, comparable to its occurrence in primary aortic dissection repair [1]. This may reflect improved surgical technique, including use of the right subclavian or axillary artery for arterial infusion protocols for brain protection and adjunctive measures [13].
In summary, aortic dissection after cardiovascular surgery can be managed with acceptable operative risks. Long-term survival is good, and need for subsequent aortic reoperation is rare.
| Appendix Variables Considered in Multivariable Analysis |
|---|
|
|
|---|
SYMPTOMS
New York Heart Association functional class (IIV), current dyspnea.
PATHOLOGY
Mitral regurgitation, aortic regurgitation, tricuspid regurgitation, pulmonary regurgitation, bicuspid aortic valve.
CARDIAC COMORBIDITY
Atrial fibrillation, complete heart block, ventricular arrhythmia, emergency cardiac surgery, number of previous cardiac operations, coronary artery disease by coronary artery territory (left main trunk, left anterior descending artery, left circumflex artery, right coronary artery).
NONCARDIAC COMORBIDITY
Treated diabetes, hypertension, history of peripheral vascular disease, presence of carotid disease, presence of chronic obstructive pulmonary disease, renal disease, history of smoking, creatinine (mg/dL), blood urea nitrogen (mg/dL), bilirubin (mg/dL), hemoglobin (g/dL), hematocrit (%), cholesterol (mg/dL), and triglyceride levels (mg/dL).
PREVIOUS OPERATIONS
Coronary artery bypass grafting, valve procedure, aneurysm repair.
EXPERIENCE
Date of surgery (date of operation from January 1, 1990).
Intraoperative Variables
PROCEDURE
Coronary artery bypass grafting, aortic valve replacement, aortic valve repair, ascending aorta procedure, aortic arch procedure, aortic valved conduit, descending aorta procedure.
SUPPORT
Aortic clamp time (min), circulatory arrest, circulatory arrest time (min), retrograde brain perfusion.
| References |
|---|
|
|
|---|
Related Article
This article has been cited by other articles:
![]() |
L.-Z. Sun, R.-D. Qi, Q. Chang, J.-M. Zhu, Y.-M. Liu, C.-T. Yu, B. Lv, J. Zheng, L.-X. Tian, and J.-G. Lu Is total arch replacement combined with stented elephant trunk implantation justified for patients with chronic Stanford type A aortic dissection? J. Thorac. Cardiovasc. Surg., October 1, 2009; 138(4): 892 - 896. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. P. Schwartz, M. Bakhos, A. Patel, S. Botkin, and S. Neragi-Miandoab Impact of pre-existing conditions, age and the length of cardiopulmonary bypass on postoperative outcome after repair of the ascending aorta and aortic arch for aortic aneurysms and dissections Interactive CardioVascular and Thoracic Surgery, October 1, 2008; 7(5): 850 - 854. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. J. Beller, M. R. Labrosse, S. Hagl, M. M. Gebhard, and M. Karck Aortic root motion remodeling after aortic valve replacement - implications for late aortic dissection Interactive CardioVascular and Thoracic Surgery, June 1, 2008; 7(3): 407 - 411. [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 |