| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2002;74:1066-1070
© 2002 The Society of Thoracic Surgeons
a Department of Cardiothoracic Surgery, Weill Medical College of Cornell University, New York, New York, USA
Accepted for publication June 5, 2002.
* Address reprint requests to Dr Girardi, 525 East 68th Street, M-424, New York, NY 10021 USA
e-mail: lngirard{at}mail.med.cornell.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and methodsResultsCommentReferences |
|---|
Methods. One hundred seventy-two patients with thoracoabdominal or descending thoracic aneurysms were operated on between July 1997 and October 2001. Forty presented with either a contained or free rupture. Three techniques were used for aortic reconstruction: clamp and sew, left heart bypass, and hypothermic circulatory arrest. Adjuncts for neurologic and renal support were used when circumstances and anatomy permitted.
Results. Seven of 40 patients died in the hospital (17.5%). Four patients died intraoperatively, all of acute myocardial infarction. Five of the seven deaths were in patients who presented in shock. Two patients (5%) experienced paraplegia, 3 (7.5%) had renal failure requiring hemodialysis, 8 (20%) required a tracheostomy, and 6 (15%) had recurrent nerve palsies. There was one stroke (2.6%). Mean diameter of ruptured aneurysms was 8.5 cm.
Conclusions. Ruptured thoracic aneurysms can be repaired with a gratifying rate of salvage. Rapid diagnosis and triage for repair is necessary to avoid progressive deterioration into shock. The incidence of myocardial infarction, and the mortality associated with this event, underscores the need for aggressive cardiac evaluation in the elective thoracic aneurysm patient. The size at rupture also emphasizes the need for earlier referral for elective aneurysm repair.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentReferences |
|---|
Although the mortality for repair of ruptured abdominal aortic aneurysms is well defined at nearly 20 times that of elective repair [6], the risk associated with repair of a ruptured thoracic aneurysm remains more obscure. A previous report documented outcome in 117 patients with ruptured DTA and TAAA [7]. However, this cohort of patients was collected during a 28-year period, included patients who ruptured during elective operations, and reported 30-day, rather than in-hospital, mortality. Adjuncts for neurologic and renal protection were not used. Two additional reports examined outcomes in patients having emergent thoracic aneurysm repair, many with the addition of methods for end-organ protection [8, 9]. The results reported by Velazquez and coworkers [8] are difficult to interpret as nearly one third of their patients had nonruptured, symptomatic aneurysms rather than ruptures. Remarkably, neither report suggests an increase in the incidence of operative mortality, paraplegia, or renal failure above that of patients having an elective operation. In contrast, our experience during a relatively short period intimated that patients with thoracic aneurysm rupture were at a significantly higher risk for poor outcome than their elective counterparts. We analyzed a more contemporary experience with ruptured thoracic aneurysms to define outcomes and risk factors associated with this highly lethal condition. An algorithm for operative decision making, including the application of end-organ protection when appropriate, is proposed.
|
Material and methods |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentReferences |
|---|
|
Aortic cross-clamping was used when control of the aorta could be obtained just proximal or distal to the left subclavian artery. All patients were given 100 U of heparin per kilogram. In a majority of cases, proximal cross-clamping with partial exsanguination and an open distal technique was used [10]. When left atrial to distal aortic bypass could be safely established, flows of 3 to 5 L/min were maintained without monitoring distal aortic pressure. Nine-French balloon perfusion and occlusion catheters (Ideas for Medicine, Clearwater, FL) provided continuous flow to the renal and visceral vessels during reconstruction of extent II aneurysms [1]. Flows were maintained at 150 mL/min per catheter. Isolated, cold renal perfusion was used for renal protection for extent III or IV aneurysms, in which renal artery reimplantation was necessary. When proximal aortic control could not be obtained, profound hypothermic circulatory arrest with nonselective retrograde cerebral perfusion was used. Intercostal artery reimplantation was performed when anatomic and hemodynamics permitted safe, timely inclusion into the reconstruction.
In the immediate postoperative period mean arterial pressure was maintained at or above 90 mm Hg. Intrathecal pressure was maintained at less than 10 mm Hg for 48 to 72 hours with intermittent drainage. The hematocrit was maintained greater than 30%.
Statistical analysis was performed using SPSS statistical software package (SPSS, Inc, Chicago, IL). Comparison between surviving and nonsurviving patients, and relative risk, was determined using Fisher’s exact test, with statistical significance being defined at p less than 0.05.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentReferences |
|---|
|
Nineteen aneurysms were atherosclerotic in nature, and seven ruptured in the setting of a chronic aortic dissection. Ten patients with aneurysmal, acute type B dissections experienced aortic rupture. Four aneurysm ruptures were secondary to infection. Of the mycotic aneurysms, three were secondary to Staphylococcus aureus, and one was secondary to Escherichia coli.
Aortic reconstruction was performed with the assistance of left atrial to distal aortic bypass in 13 patients, and a simple clamp and sew technique was used in 20 patients. Mean time of partial bypass was 23 minutes. Seven patients required PHCA. In the 33 patients having repair without PHCA, control of the aorta was obtained proximal to the left subclavian artery in 11 patients and distal to the left subclavian artery in 22 cases. Extent of reconstruction is listed in Table 1 according to a previously described segmental division of the thoracoabdominal aorta [4]. Mean aortic cross-clamp time for DTA and TAAA repair was 27 and 37 minutes, respectively. Mean duration of PHCA was 18 minutes.
Twenty patients had a cerebrospinal fluid drain before proceeding with aneurysm repair whereas 20 patients underwent repair without spinal fluid drainage. Six of 7 patients repaired under PHCA did not receive drains, and 14 patients were too unstable to permit drain insertion. Nine patients had a mean of two sets of intercostal arteries reimplanted during aortic reconstruction. Mean core temperature of the non-PHCA patients during aortic cross-clamping was 33.5°C.
A number of procedures were required in addition to primary aortic reconstruction. All 7 patients having repair under PHCA had distal hemiarch reconstruction. Three patients required aortorenal bypass, 1 aortosubclavian bypass, and 1 aorta to superior mesenteric artery bypass. Three patients having ruptured mycotic aneurysms had serratus anterior muscle flaps taken down at the end of the procedure. The muscle flap was wrapped around the aortic graft after being brought through the third intercostal space. The remaining mycotic rupture was treated by ascending to descending aortic bypass and descending aortic exclusion [11].
Postoperative complications for this entire cohort of patients are listed in Table 2. Four patients (10%) experienced fatal MI. All 4 had acute ST-segment elevation that subsequently deteriorated into ventricular fibrillation. Three of the 4 occurred intraoperatively. In these 3, intraoperative transesophageal echocardiography documented significant changes in left ventricular function preceding and during the infarction. Three patients (7.5%) required hemodialysis before discharge. One patient had a creatinine of 3.5 mg/dl at the time of presentation. His need for dialysis resolved before discharge. The second patient with renal failure had a creatinine of 5.8 on presentation and ultimately required permanent hemodialysis. Eight patients (20%) required temporary tracheostomy after operation, and 6 patients (15%) had permanent left recurrent nerve palsies. Two patients experienced paraplegia after aneurysm repair (5%). Paraparesis after ruptured aneurysm repair was not seen. There was one perioperative stroke (2.5%). No patient required reexploration for bleeding.
|
|
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentReferences |
|---|
In both ruptured TA and ruptured abdominal aortic aneurysms, shock is a significant risk factor for operative mortality [6, 7]. Shock eliminates the ability to use end-organ protection and potentiates an inflammatory cascade that contributes heavily to multiorgan dysfunction. Cardiac dysfunction, specifically MI, is highly associated with hemorrhagic shock in the setting of a ruptured TA. Three of four MIs occurred in patients in shock and were ultimately responsible for 57% of the deaths in this series. Myocardial infarction occurred three times intraoperatively, and once postoperatively, and was uniformly fatal. Crawford and colleagues [7] reported a similar pattern of myocardial ischemia when caring for patients with ruptured TA. Seventy-five percent of the deaths in their series were cardiac in origin. A majority occurred in hypotensive patients or those having a preoperative cardiac arrest. Although little can be done to preemptively treat significant coronary artery disease in the presence of thoracic aneurysm rupture, all patients in our practice receive a thorough cardiac evaluation before elective aneurysm repair. Patients with significant coronary stenoses undergo revascularization before aneurysm repair. A period of 4 to 6 weeks is usually necessary for the patient to adequately recover from angioplasty or bypass operation. During this time aggressive blood pressure control and ß-blocking medication is necessary to minimize further aortic expansion. Using this approach, we have not had an MI in any patient undergoing elective TA repair, despite advanced age and low ejection fractions. Additionally, no patient has experienced a rupture or sudden death while waiting for stabilization of their new coronary circulation.
As the flow chart in Figure 1 illustrates, patients with free rupture and hemodynamic compromise should be taken immediately to the operating room for repair. Rapid proximal control of the aorta is the most critical maneuver and can be challenging in the setting of ongoing exsanguination, especially if there is no time to establish single-lung ventilation. Additional difficulty may be encountered if the aorta is dilated at the level of the left subclavian artery, making cross-clamp control of the distal arch even more difficult. In this instance, the ligamentum arteriosum is rapidly divided, creating adequate space for a cross-clamp just beyond the left common carotid artery. Clamping the arch in this fashion is not associated with an increased risk of stroke [12]. The recurrent nerve is often contused or transected during this maneuver, and unilateral vocal cord paresis can be expected. If proximal control cannot be obtained, digital control of the rupture, femoral-femoral cardiopulmonary bypass, and PHCA may be lifesaving.
If the rupture is contained and the patient is stable, adjuncts for spinal cord and renal protection may be used. If repair can be achieved with an anticipated cross-clamp time of less than 30 minutes, then a simple clamp and sew technique, with spinal drainage, is used. If a more complex repair is contemplated, as in most cases of acute or chronic aortic dissection or an extent II aneurysm, then left heart bypass is instituted. As in the case of free rupture, inability to obtain adequate proximal control of the aorta mandates the use of PHCA.
We believe the best method to reduce the mortality of thoracic aneurysm operation is avoiding rupture with an elective operation. The mortality for elective DTA and TAAA repair has continued to fall during the last decade. In the elective setting, a perioperative mortality of 5% or less can be expected, with similarly low rates of permanent neurologic damage [1, 13–15]. Our mortality for a ruptured TA, 17.5%, is more than five times our mortality for elective aneurysm repair and is similar to the mortality achieved for a comparable group of patients reported on by von Segesser and colleagues [9] and Velazquez and associates [8] (15% and 13%, respectively). The latter two reports suggest that the mortality for repair of ruptured TA is not statistically greater than that for patients having elective aneurysm repair. von Segesser and associates [9] suggest that similar cross-clamp times between elective and emergent cases can account for the similar mortality between these heterogeneous groups. They also suggest that hemodynamic stability at the time of repair contributes to surgical success for ruptured patients, although there is no report of the number of patients with hemodynamic instability or shock at the time of operation. Velazquez and coworkers [8] attribute the success of their repaired ruptured aneurysms to the use of adjuncts for end-organ protection and the accompanying reduction in the global systemic effects of reperfusion injury. Although we agree that minimizing end-organ ischemic time is critical to successful outcome in any TA repair, we concur with Cambria [16] that the high mortality for elective aneurysm repair in this report may skew the anticipated outcome of patients presenting with aneurysm rupture. We believe that the mortality for repair of TAs can be expected to be significantly greater than with elective repair, and are gratified by the surprisingly high rate of salvage that can be obtained under such dire circumstances.
The negative impact of aortic rupture on survival is even more pronounced in ruptured abdominal aortic aneurysms, in which the perioperative mortality is nearly 50% compared with an elective, open repair mortality of 2% [6]. Continued dialogue with the physicians caring for patients with aortic disease is clearly needed. Despite known aortic disease, the mean diameter of all ruptured TA in this study, 8.2 cm, was nearly 2 cm greater than for our patients undergoing elective repair during this same period. This is even more startling when considering that greater than 50% of these patients had known chronic aortic dissections or had undergone previous aneurysm repair. A more thorough understanding of the necessity for serial radiographic examination, and the criterion for elective aneurysm repair, may reduce the number of patients presenting with rupture.
The risk of death, aortic rupture, or dissection increases to 6.9% per year once the aortic diameter is 6 cm or greater [17]. Patients with chronic dissections or chronic obstructive pulmonary disease also have a greater propensity toward rupture at smaller diameters [18]. Once patients approach this critical diameter anywhere along the thoracoabdominal aorta, a thorough multisystem evaluation is warranted in preparation for open repair. With modern surgical techniques and improvements in critical care in the postoperative period, few patients would be excluded from traditional open aneurysm repair because of organ insufficiency. An overwhelming majority could expect an excellent surgical result. Perhaps the small number of patients deemed poor surgical candidates, especially those with severe chronic obstructive pulmonary disease or known renal failure, would have anatomy suitable for elective endovascular repair.
Patients with ruptured thoracic aneurysms can be salvaged if rapid diagnosis and repair can be performed. Expeditious proximal control and reconstruction, before the development of hemorrhagic shock, is critical in minimizing irreversible myocardial ischemia. Aggressive intensive care unit support will allow many of these patients to survive to a satisfactory discharge.
|
References |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  H.-S. Hofmann, H. Kroll, C. Kunze, and H. Bromber Should Patients With Contained Rupture of a Descending Aortic Aneurysm Only Receive Unilateral Artificial Ventilation? Case Report of a Death During an Operation Vascular and Endovascular Surgery, March 1, 2008; 42(1): 82 - 84. [Abstract] [PDF]
|
|
|
|
 |
|
 L. N. Girardi, K. H. Krieger, C. A. Mack, L. Y. Lee, A. J. Tortolani, and O. W. Isom Does Cross-Clamping the Arch Increase the Risk of Descending Thoracic and Thoracoabdominal Aneurysm Repair? Ann. Thorac. Surg., January 1, 2005; 79(1): 133 - 137. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Morishita, Y. Kurimoto, N. Kawaharada, J. Fukada, Y. Hachiro, Y. Fujisawa, and T. Abe Descending Thoracic Aortic Rupture: Role of Endovascular Stent-Grafting Ann. Thorac. Surg., November 1, 2004; 78(5): 1630 - 1634. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. J. Drenth, E. L. G. Verhoeven, T. R. Prins, T. W. Waterbolk, and P. W. Boonstra Relocation of supra-aortic vessels to facilitate endovascular treatment of a ruptured aortic arch aneurysm J. Thorac. Cardiovasc. Surg., October 1, 2003; 126(4): 1184 - 1185. [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 |
