|
|
||||||||
Ann Thorac Surg 2003;76:1951-1956
© 2003 The Society of Thoracic Surgeons
a Section of Cardiovascular Surgery, School of Medicine, Keio University, Tokyo, Japan
Accepted for publication June 6, 2003.
* Address reprint requests to Dr Ueda, Section of Cardiovascular Surgery, School of Medicine, Keio University, 35 Shinanomachi, Shinjuku-ku, 160-8582 Tokyo, Japan.
e-mail: uedatosh{at}kmh.gr.jp
| Abstract |
|---|
|
|
|---|
METHODS: Between July 1993 and November 2001, 103 patients (mean age 65 ± 11 years, 26 women, 35 dissections) underwent total arch replacement through a median sternotomy using a branched arch graft with selective cerebral perfusion. Eighteen operations including 14 acute dissections were performed on an emergency basis. Concomitant procedures were root replacement in 5 patients, mitral valve replacement in 1, coronary artery bypass in 14, and open endovascular stent-graft in 9. The average time (minutes) for bypass, aortic cross-clamp, selective cerebral perfusion, and distal arrest were respectively 273 ± 79, 163 ± 54, 145 ± 36, and 69 ± 22.
RESULTS: Mechanical heart support was necessary in 3 patients. Stroke occurred in 9 patients, transient neurologic dysfunction in 7, and paraplegia/paraparesis in 4. The only independent determinant for postoperative stroke was a history of stroke (odds ratio 16.3, 95% confidence interval: 2.8 to 93.8). Thirty-one patients required ventilator support for more than 5 days. Hemodialysis was needed in 5 patients. Sternal infection or mediastinitis occurred in 6 patients. The in-hospital mortality was 12% (12 of 103). The actuarial survival rate at 1 year was 83%, and was 67% at 5 years. For the 1-year mortality independent determinants were emergency surgery (odds ratio 5.3, 95% confidence interval: 1.6 to 17.9) and age 75 years or older (odds ratio 4.0, 95% confidence interval: 1.1 to 13.9).
CONCLUSIONS: Total arch replacement using a branched arch graft with selective antegrade cerebral perfusion has a favorable 1-year mortality rate except for patients undergoing emergency surgery and for elderly patients.
| Introduction |
|---|
|
|
|---|
Kazui and colleagues [7] have reported outcomes after total arch replacement using a branched arch graft with selective cerebral perfusion. They found that hospital mortality ran at 13% but the survival rate at 1 year after surgery was 83%. In the study by Estrera and colleagues [4] the survival curve in patients undergoing ascending or aortic arch operation or both with hypothermic circulatory arrest and retrograde cerebral perfusion also showed a steep fall over the year after the operation. Some patients who survived the operation died soon after discharge. Recent studies on the natural history of aortic aneurysms also look at the estimated yearly rupture rate [8, 9]. The present study aims to determine the risk factors associated with mortality within 1 year postoperatively in patients undergoing total arch replacement through a median sternotomy using a branched arch graft with selective cerebral perfusion.
| Material and methods |
|---|
|
|
|---|
|
A history of stroke was noted in 8 patients and 2 other patients had experienced a transient ischemic attack. One patient had spinocerebellar degenerative disease. Coronary artery disease was noted preoperatively in 17 patients. Five patients had a history of chronic obstructive pulmonary disease. Renal dysfunction (glomerular filtration rate < 40 mg · mL-1 · min-1 or serum creatinine level > 2.0 mg/dL) was noted in 14 patients but no patient needed preoperative hemodialysis.
Operative procedures
Details of the surgical techniques and selective cerebral perfusion are described elsewhere [6, 10]. Median sternotomy was performed in all patients. The left sternocleidomastoid muscle was incised to access the arch vessels. The cardiopulmonary bypass circuit included a centrifugal pump and a membrane oxygenator together with a heat exchanger. The arterial line was bifurcated for the femoral artery and for antegrade perfusion through the ascending aorta or through a branch of the graft. A second arterial line equipped with an air filter was used for selective perfusion of the arch vessels using a roller pump. The end of the line was trifurcated and was then connected to the perfusion catheters, which had a balloon and a pressure monitor line in their tips (15F for the braciocephalic artery, 12F for the left common carotid and the left subclavian artery).
One arterial line was always placed in the femoral artery. When malperfusion was anticipated in patients with aortic dissection or when the distal aorta was found to contain mural thrombi, the main perfusion route was instituted through a cannula in the ascending aorta or the right subclavian artery (n = 66). A two-stage venous cannula was used in most patients. After institution of total cardiopulmonary bypass the patient was cooled. The left vent catheter was then inserted through the right superior pulmonary vein. Myocardial protection was based on antegrade perfusion of either crystalloid or blood cardioplegic solution. Distal anastomosis of the coronary artery bypass and the aortic root procedures were performed during the period of cooling when cross-clamps on the ascending aorta could be applied.
Circulation was arrested at a rectal temperature of below 23°C and the aortic arch was opened and the three arch vessels severed in a single cut. A perfusion catheter was inserted into each arch vessel and was gently secured using a tourniquet after the balloon had been inflated. The perfusion flow rate was maintained initially at 10 mL · min-1 · kg-1 and then adjusted to ensure the perfusion pressure measured at the tip of the catheter in the left common carotid artery remained between 40 and 60 mm Hg.
Woven Dacron vascular grafts sealed with collagen (Hemashield; Meadox Medical, Oakland, NJ) were used. Before 1997 a branched arch graft was constructed in the operating room; since then the graft has been commercially available. Distal aortic anastomosis was first performed. Upon completion of the anastomosis any debris or air in the descending aorta was evacuated by retrograde perfusion through the femoral cannula and antegrade perfusion was established through a branch graft. In the earlier cases anastomosis and unclamping of the left subclavian artery, the proximal aorta, the left common carotid artery, and the braciocephalic artery were performed in that order. Rewarming was started during anastomosis of the braciocephalic artery. From April 1999 in order to prevent mismatches between brain metabolism and temperature reattachment of the arch vessels was performed before the proximal aortic anastomosis and rewarming began only after cessation of selective cerebral perfusion. The flow rate of each arch vessel during selective cerebral perfusion has also been monitored since 1999 with a transit type ultrasonic flow meter.
Concomitant procedures are shown in Table 2: aortic root replacement in 5 patients, mitral valve replacement in 1, coronary artery bypass in 14, and left upper lung resection in 1 patient. In patients with acute dissection, gelatin-resorcinol-formaldehyde glue was used as an adjunct to secure anastomosis. In 3 patients whose left vertebral artery originated separately from the aortic arch, the saphenous vein graft was interposed in an end-to-side manner between the artery and the branch graft reattachment to the left subclavian artery. Distal aortic anastomosis was performed by the elephant trunk technique in 24 patients. Between October 1999 and June 2000 the endovascular open stent-graft technique was used in 9 patients to eliminate distal aortic diseases, after the report by Kato and colleagues [11]. However we gave up this technique as likely to cause spinal damage [12].
|
Data analysis
All statistical analyses were performed using the SPSS statistical software program (SPSS, Chicago, IL). The Kaplan-Meier method was used to calculate the cumulative survival rate. Continuous data were expressed as mean ± 1 standard deviation. Univariate analysis was performed using Fisher's two-tailed exact test. All variables showing a p value less than 0.1 on the exact test were included in the multivariate analysis. Independent risk factors for postoperative neurologic stroke and any death within a year of surgery were examined using a stepwise multiple logistic regression analysis. All p values less than 0.05 were taken as significant.
| Results |
|---|
|
|
|---|
|
|
Mortality after surgery
The 30-day mortality was 7 of 103 (Table 5).
The causes of these 7 deaths were low cardiac output in 3 patients, respiratory failure in 2, mediastinitis in 1, and laryngospasm in 1. Five other patients died in hospital, 1 each from respiratory failure, aspiration, mediastinitis, multiple organ failure, and accident. Of the 91 operative survivors, 6 died within 1 year of surgery. Three of these 6 patients were severely exhausted at the time of discharge (cachexic) although they had not suffered any postoperative neurologic complication. They could not eat well at home. Two of these 3 patients died of pneumonia and the third died probably as a result of renal failure. One patient died 2 months after discharge of distal aortic rupture. His planned second operation had been postponed in view of his poor recovery from total arch replacement. The causes of the remaining two 1-year deaths were subarachnoid hemorrhage and peritonitis after resection of an ovarian tumor. Thirteen additional deaths occurred more than 1 year after discharge: 1 patient died as a consequence of reoperation due to endocarditis, 2 patients died of aortic rupture, and 1 patient suffered sudden death. The causes of death in the remaining 9 patients were pneumonia in 3, renal failure in 2, cervical osteomyelitis in 1, and unknown in 3. The Kaplan-Meier actuarial survival rate was 83% at 1 year and 67% at 5 years (Fig 1).
|
|
|
| Comment |
|---|
|
|
|---|
The long distal arrest time in the present study is attributed to the difficulty of distal anastomosis. Because the suture line was often very deep in the left thorax we preferred meticulous mattress sutures to secure distal anastomosis. Although the incidence of postoperative spinal damage was not low (4%), each patient who suffered paraplegia/paraparesis had some cause (stated in the Results section) other than a long distal arrest time.
Neurologic problems were by far the most serious postoperative complications after aortic arch repair. Svensson and colleagues [13] recently reported the outcome of a multimodal protocol for brain protection. In their study in which either hypothermic circulatory arrest with retrograde cerebral perfusion or selective cerebral perfusion was used stroke occurred in only 2.0% of 403 patients undergoing ascending or arch aortic operations. They conclude that the benefit of antegrade or retrograde perfusion is unproven. Increasing evidence shows however that occurrence of cerebral complications associated with hypothermic circulatory arrest is time-dependent irrespective of the use of retrograde cerebral perfusion [2, 3, 14]. In contrast with selective cerebral perfusion the duration of brain protection was not a determinant of cerebral complication [5, 7, 10]. The independent determinants of permanent neurologic dysfunction associated with selective cerebral perfusion are old cerebral infarct, pump time [7], urgency status [5], and in the current study a history of stroke.
Most of our patients had distal aortic arch lesions caused by severe arteriosclerosis. Also the orifice of each arch vessel usually had a dirty atheroma. We believe that such lesions should be extirpated using the branch grafts. Both distal anastomosis deep in the left thorax and reattachment of the diseased arch vessel can take long times. We therefore preferred selective cerebral perfusion to hypothermic circulatory arrest for cerebral protection during total arch replacement using a branched graft. Nine patients in our study had a history of stroke and 4 of these had a further stroke postoperatively. Cerebrovascular occlusive disease was found to exist preoperatively in all 4 of these patients. A history of stroke is still a risk factor for postoperative stroke [10]. From mid 1999 we altered the timing of rewarming. To prevent mismatches between brain metabolism and temperature, rewarming was begun only after cessation of selective cerebral perfusion. The number of patients operated on by this new technique is still too small for statistical evaluation.
Axillary/subclavian artery perfusion has been recommended to prevent atheroma emboli [15]. The present study found no difference in the incidence of stroke between patients undergoing femoral artery perfusion and patients undergoing subclavian artery perfusion. We maintain that retrograde perfusion through the femoral artery is a safe and useful technique for appropriately chosen patients.
The in-hospital mortality rate of 12% in the present study was similar to the rate of 13% reported by Kazui and colleagues [7]. The survival curves after aortic arch surgery show a steep fall over the year after the operation, especially among patients undergoing total arch replacement [4, 7]. The 1-year survival rate was as low as 83% in both our study and that of Kazui and colleagues. Recent studies on the natural history of thoracic aortic aneurysms referred to the yearly death, dissection, or rupture rate [8, 9]. We therefore focus on the 1-year mortality.
One of the two independent determinants of 1-year mortality in the current study is emergency surgery, which is a known risk factor for hospital death in patients undergoing aortic arch surgery [3, 5, 7]. Kazui and colleagues [7] suggested that in-hospital mortality after total arch replacement is determined principally by the preoperative state of the patients. This explains their recent conclusion that total arch replacement should not be performed routinely in patients with acute type-A aortic dissection [16]. Unlike degenerative aortic aneurysm the extent of aortic resection in acute type-A aortic dissection often depends on the surgeon's experience [17, 18]. The present results support the conclusion of Kazui and colleagues [16].
Another determinant of 1-year mortality is age 75 years or older. Elderly patients often did not recover well after total arch replacement even if they had been healthy preoperatively. Furthermore some of them became systemically exhausted and died of pulmonary infection or other unknown causes after discharge. Total arch replacement using the present technique clearly imposes great surgical stress on patients, to which the elderly are less resilient. Similarly when hypothermic circulatory arrest with or without retrograde cerebral perfusion was used for brain protection, age was reported to be an independent risk factor of hospital mortality and of poor long-term survival [3, 4].
Elefteriades [19] concludes that a patient having an aorta that has reached 6 cm maximal diameter should expect rupture or dissection within 1 year and an annual death rate of 14%. Excluding our 36 patients having the risk factors for 1-year mortality (18 patients undergoing emergency surgery and 18 patients aged 75 or older), we find a 1-year mortality rate of 10% (7 of 69). We therefore believe that our current technique of total arch replacement is clinically acceptable for patients aged 74 years or less who undergo the operation as scheduled.
| Appendix |
|---|
Age more than = 65 years, more than = 70 years, more than = 75 years (pd = 0.082); sex (pd = 0.069); Marfan syndrome; prior aortic surgery [proximal aorta, distal aorta]; history of stroke (ps = 0.005); coronary artery disease; renal dysfunction; chronic obstructive lung disease; emergency surgery (pd = 0.015); preoperative shock (pd = 0.078); aortic dissection [acute (pd = 0.067), chronic]; aortic root replacement; coronary artery bypass grafting; elephant trunk technique; open endovascular stent grafting; arterial perfusion through the femoral artery; pump time more than 240 minutes (ps = 0.067), more than 300 minutes; aortic cross-clamp time more than 150 minutes (ps = 0.078), more than 180 minutes (ps = 0.056), more than 210 minutes (ps = 0.056); distal circulatory arrest time more than 60 minutes, more than 90 minutes; blood transfusion more than = 20 units, more than = 30 units, more than = 40 units, more than = 50 units (pd = 0.095); platelet transfusion more than = 20 units (pd = 0.066), more than = 30 units, more than = 40 units.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
P. G. Malvindi, G. Scrascia, and N. Vitale Is unilateral antegrade cerebral perfusion equivalent to bilateral cerebral perfusion for patients undergoing aortic arch surgery? Interactive CardioVascular and Thoracic Surgery, October 1, 2008; 7(5): 891 - 897. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Kawasaki, K. Sugimoto, T. Taniguchi, M. Yamaguchi, M. Fujii, K. Sugimura, and Y. Okita Endovascular Treatment for Visceral Vessel Complication After Branched Graft Replacement: Initial Results Am. J. Roentgenol., October 1, 2008; 191(4): 1175 - 1181. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. Numan, H. Arbatli, W. Bruszewski, and M. Cikirikcioglu Total endovascular aortic arch reconstruction via fenestration in situ with cerebral circulatory support: an acute experimental study Interactive CardioVascular and Thoracic Surgery, August 1, 2008; 7(4): 535 - 538. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Antona, P. Vanelli, M. Petulla, G. Gelpi, P. Danna, M. Lemma, and L. Inglese Hybrid Technique for Total Arch Repair: Aortic Neck Reshaping for Endovascular-Graft Fixation Ann. Thorac. Surg., March 1, 2007; 83(3): 1158 - 1161. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Suzuki, T. Kazui, A. H. M. Bashar, K. Yamashita, H. Terada, N. Washiyama, and T. Suzuki Total Aortic Arch Replacement in Patients With Arch Vessel Anomalies Ann. Thorac. Surg., June 1, 2006; 81(6): 2079 - 2083. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Della Corte, M. Scardone, G. Romano, C. Amarelli, A. Biondi, L. S. De Santo, M. De Feo, G. Nappi, and M. Cotrufo Aortic Arch Surgery: Thoracoabdominal Perfusion During Antegrade Cerebral Perfusion May Reduce Postoperative Morbidity Ann. Thorac. Surg., April 1, 2006; 81(4): 1358 - 1364. [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 |