| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2003;76:1477-1484
© 2003 The Society of Thoracic Surgeons
a Second Department of Surgery, Faculty of Medicine, University of the Ryukyus, Okinawa, Japan
Accepted for publication May 6, 2003.
* Address reprints to Dr Kuniyoshi, Second Department of Surgery, Faculty of Medicine, University of the Ryukyus, 207 Uehara Nishihara cho, Okinawa 903-0215, Japan
e-mail: kuni9244{at}med.u-ryukyu.ac.jp
Abstract
BACKGROUND: We present data showing the impact of sequential multisegmental aortic clamping accompanied by reimplantation of as many segmental arteries as possible on the prevention of postoperative paraplegia or paraparesis during thoracoabdominal aortic graft replacement.
METHODS: Since 1987 we have performed graft replacements in 51 individuals undergoing thoracoabdominal aortic surgery using the technique of normothermic partial bypass with sequential multisegmental aortic clamping. The procedure was performed emergently in 10 patients and electively in 41 patients. The patients ranged in age from 22 to 82 years (mean, 57.6 ± 13.8 years). Indications for surgery included dissecting thoracoabdominal aortic aneurysm (n = 19) and nondissecting thoracoabdominal aortic aneurysm (n = 32). The extent of aneurysm was Crawford type I in 19 patients, type II in 7 patients, type III in 12 patients, and type IV in 13 patients. Along the entire extent of aneurysm to be replaced, we reimplanted as many of the patent segmental arteries as feasible.
RESULTS: Five patients died during hospitalization, for an in-hospital mortality rate of 9.8%. The number of aortic clampings per patient ranged from one to five (median, three). A total of 124 segmental arteries were reimplanted in 44 (86.3%) of 51 patients. Of the 124 arteries, 90 (72.6%) were distributed between T9 and L2. Postoperative paraplegia or paraparesis did not develop in any of the patients.
CONCLUSIONS: Our results demonstrate that extensive reimplantation of segmental arteries using sequential multisegmental aortic clamping, accompanied by adequate intraoperative distal aortic perfusion, is effective in preventing spinal cord ischemia.
To treat thoracoabdominal aortic aneurysms (TAAA) successfully, intraoperative protection of the visceral organs and spinal cord is essential. However, complications such as postoperative renal failure, hepatic failure, and paraplegia may still occur. Postoperative paraplegia has remained the most severe complication, and to date there has been no definitive means of preventing this outcome [1–11]. Intraoperative or postoperative ischemia of the spinal cord is generally accepted as one of the main causes of this complication.
To maintain spinal cord circulation intraoperatively and postoperatively, various surgical methods and adjuncts have been reported, such as sequential clamping and segmental repair [12, 13]. We applied the sequential clamping and segmental repair method to the whole extent of the aneurysm, while maintaining adequate distal aortic perfusion by means of a normothermic partial bypass. The aneurysm was divided into segments that did not exceed approximately three vertebral bodies in length. In addition, we reimplanted as many segmental arteries with patent ostia as possible [14]. Using these operative techniques, we have operated on 51 consecutive patients without any postoperative neurologic abnormalities. In this report, we present our clinical experience with TAAA surgery.
Patients and methods
Between 1987 and December 2002, we treated 51 consecutive patients with TAAA (age range, 22 to 82 years; average, 57.6 ± 13.8 years; 34 men and 17 women). The 51 patients are enumerated in Table 1. The patients' aneurysms consisted of dissecting (n = 19) and nondissecting (n = 32) types. The TAAA extent was characterized using the classification proposed by Crawford and colleagues [1]: type I (n = 19), involvement of most of the descending thoracic and upper abdominal aorta; type II (n = 7), involvement of most of the descending thoracic aorta and most or all of the abdominal aorta; type III (n = 12), involvement of the distal descending thoracic aorta and varying segments of the abdominal aorta; and type IV (n = 13), involvement of the abdominal aorta including the segment from which the visceral vessels arose. Ten patients (19.6%) underwent emergency operations owing to aneurysmal rupture or impending rupture associated with severe thoracolumbar pain. Before TAAA surgery 13 patients had undergone one of the following procedures: coronary artery bypass grafting (n = 3); graft replacement of the upper part of the thoracic descending aorta (n = 4); Bentall's operation plus graft replacement from the ascending aorta to the middle of the thoracic descending aorta (n = 2); graft replacement from the ascending aorta and the aortic arch plus coronary artery bypass grafting (n = 1); graft replacement from the distal arch to the thoracic descending aorta (n = 1); two previous operations consisting of Bentall's operation and graft replacement of the thoracic descending aorta (n = 1); and two previous operations consisting of graft replacement of the ascending aorta and the thoracic descending aorta (n = 1). Ten patients had a history of aortic surgery proximal to the present aortic aneurysm. The previous surgery extended to the distal arch in 1 patient, to T6 (6th intercostal artery), T7, T8, T9, and T11 in 1 patient each, and to T10 in 4 patients. Of these patients, one had undergone reconstruction of a segmental artery of T8. The period between the latest operation and the present operation was 1 month to 12.2 years, with an average of 2.65 ± 4.1 years.
|
Preoperatively, agents that protect against spinal cord ischemia or cerebrospinal fluid drainage were not used. Patients were anesthetized and intubated using a double-lumen endotracheal tube. Patients were placed in a right decubitus position with the lower half of the body rotated posteriorly to facilitate right femoral vessel cannulation. The right femoral artery and vein were used as inflow and outflow sites because the long venous drainage cannula was easier to insert into the right atrium from the right femoral vein than from the left femoral vein. The surgical approach to the aneurysm was carried out through Stony's incision. The entire aneurysm was exposed, with tapes applied for multisegmental aortic clamping at intervals of approximately every three vertebral lengths. A full dose of heparin (3 mg/kg) was administered. Adjuncts for aortic clamping consisted of normothermic partial femoral vein–femoral artery bypass (n = 45) or pulmonary artery–femoral artery bypass (n = 6). There were no complications such as lung bleeding during pulmonary artery–femoral artery bypass.
The cardiopulmonary bypass circuit was an open system, and included a cardiotomy reservoir, an oxygenator, and a heat exchanger. The average maximal perfusion flow rate was 3.02 ± 0.91 L/min at the start of partial cardiopulmonary bypass, and the mean pressure of the distal aorta monitored from the femoral artery pressure was maintained at 87.8 ± 17.9 mm Hg. The average aortic clamp time ranged from 29 to 211 minutes (average, 121.1 ± 47.1 minutes). The minimum clamp time was noted in case 5, a patient with Crawford type III TAAA, whose aneurysm extended from TV8 (8th thoracic vertebra) to above the celiac artery, with occluded intercostal arteries and a patent lumbar artery present at LV1 (1st lumbar vertebra). In this case graft replacement of the aneurysm was completed in 29 minutes, with L1 (1st lumbar vertebra) reimplantation by beveled anastomosis. The maximum clamp time was noted in case 42, a patient who had undergone two previous graft replacement procedures on the ascending aorta to the level of TV12. Rectal temperature, which was maintained by using a heat exchanger, ranged from 32° to 36.5°C (mean, 35.4° ± 1.1°C).
We have used somatosensory evoked potentials (SEP) since 1989 and motor evoked potentials (MEP) since 2000 to monitor spinal cord ischemia, and in the current study SEP was used in 46 cases and MEP in 10 cases. During the procedure patent segmental arteries were preserved using a tourniquet method until completion of graft replacement (Fig 1) . [14]. Measurements of SEP and MEP were carried out at every stage of reimplantation, and a final decision was made whether to further add the implants at the completion of graft replacement. Preserved segmental arteries were reimplanted when MEP amplitude decreased to below 25% of baseline or when SEP amplitude decreased to below 50% of baseline. Otherwise, they were ligated or oversewn. There were two cases (patients 35 and 51) that required additional reimplantation of segmental arteries because of decreases in SEP or MEP. The L2 in case 35 and T9, T10, and T11 in case 51 were reimplanted, and monitored evoked potentials were restored.
|
|
The final step in the surgical procedure was to complete the distal anastomosis of the aortic graft, which was performed with either the terminal aorta or with both iliac arteries using a Y graft. Single aortic clamping was applied in the patients who underwent emergency surgery because of aneurysmal rupture (n = 4) or for short-segmented aneurysms (n = 3). The postoperative neurologic examination was performed by surgeons and intensive care unit doctors. Because all patients awoke from anesthesia and were able to follow commands, including those who died in the early postoperative period, we could readily carry out somatomotor and somatosensory nerve examinations.
Statistical analysis
The data were analyzed using StatView 5.0 (SAS Institute, Cary, NC) statistical software package. Data are expressed as mean ± standard deviation. The cumulative survival rate was estimated by the Kaplan-Meier method. Differences in the cumulative survival rate between the dissection and nondissection groups were determined by the Mantel-Cox test.
Results
Five patients died in-hospital, for an overall mortality rate of 9.8%. These 5 patients were categorized as Crawford type I in 1, type II in 2, type III in 1, and type IV in 1. Two of the 5 deaths were in patients who underwent emergency surgery for ruptured aneurysms with shock, and the other 3 were in the elective group, giving mortality rates of 20.0% and 7.3%, respectively. The causes of early operative death were multiple organ failure in 1, peritonitis in 1, respiratory failure in 2, and pulmonary hemorrhage in 1. Specific details in each case are as follows: In case 4 with ruptured Crawford type I TAAA, an emergency operation was performed, but postoperatively shock developed in the patient and heart failure ensued. The patient died with multiple organ failure 21 days after surgery. In case 36 emergency laparotomy and resection of the small intestine was required on the fifth postoperative day because of perforation of the jejunum, and the patient subsequently had peritonitis and died on the 26th postoperative day. In case 40 with chronic dissecting aneurysm, the patient succumbed after developing respiratory failure secondary to heart failure. Case 49 suffered postoperative hepatic and respiratory failure after multiple blood transfusions associated with an increase in serum bilirubin to values exceeding 40 mg/dL. Case 42 was a patient with Marfan's syndrome who died of pulmonary hemorrhage after Bentall's operation, which was performed because of sudden onset of aortic valve regurgitation after the TAAA surgery.
The number of sequential distal aortic clampings per patient was one to five, with a median number of three clampings (Table 2).
|
|
The patients were followed until December 2002, and follow-up was completed in 100% of individuals. The mean follow-up period was 6.78 ± 4.4 years with a maximum of 15.5 years. In the late postoperative period 9 additional deaths occurred. Two patients died of rupture of a newly formed dissecting aneurysm, 1 of pneumonia, 1 of a ruptured residual aneurysm, and 1 of a rupture of the reconstructed intercostal arteries. In case 34, a patient who had undergone reimplantation of T8, T9, T10, and T11 en bloc with the carrel patch technique, the aortic wall at the juncture of these intercostal arteries dilated and formed an aneurysm. The patient died of rupture of this aneurysm 9.8 years after surgery. An additional patient died at reoperation for pseudoaneurysm after TAAA surgery, 1 died of arch aneurysm, and 2 died of ischemic heart failure. The survival rates of the patients (excluding in-hospital deaths) were as follows: 93.1% at 2 years after operation, 87.6% at 5 years, 74.8% at 10 years, and 56.6% at 14 years (Fig 4). There was no significant difference in late survival rates between dissecting and nondissecting groups (Fig 5).
|
|
The mechanism responsible for causing postoperative paraplegia has remained controversial, although it is generally thought that spinal cord ischemia is principally responsible. For this reason, intraoperative and postoperative maintenance of adequate circulation to the spinal cord is exceedingly important in preventing postoperative neurologic injuries. Svensson and associates [2] emphasized that distal perfusion pressure should be maintained above 60 mm Hg and that the flow rate should also be kept above 60 mL · min-1 · kg-1. By increasing perfusion pressure and volume, blood supply to the aortic segment at risk will increase indirectly. Ueda and colleagues [15] reported selective direct perfusion of segmental arteries for maintenance of spinal cord blood flow during opening of the aneurysm.
The existence of the collateral vessels for each segmental artery can be confirmed from intercostal or lumbar arteriograms, and by intraoperative observation of ostial back bleeding from the segmental arteries even within the clamped aortic segment. Figure 6 (case 46) shows a postoperative selective arteriogram of the reimplanted left side 11th intercostal artery, which supplies branches to the anterior spinal artery and collaterals to the 10th intercostal artery. From the findings of postoperative angiography of implanted segmental arteries, it is suspected that collaterals of segmental arteries supply blood flow for a distance of one to two vertebral bodies proximally and distally. Therefore, the maximum extent of clamped and excluded aortic segments is estimated to be approximately three vertebral bodies in length. Reimplanted segmental arteries can in turn function as collaterals to the next aortic segment. At least one segmental artery per aortic segment should be reimplanted to ensure the presence of a collateral chain among segmental arteries. In our series the distance between two reimplanted segmental arteries ranged from one to four vertebral bodies. Also, the multisegmental aortic clamp procedure that we used required a shorter interval and more aortic clamps compared with other reports [10], and we reimplanted segmental arteries that existed in areas other than between T9 to L2 as well.
|
During the follow-up period 1 patient died of aneurysm formation at the segmental artery reimplantation site at which four consecutive segmental arteries were reimplanted en bloc using the carrel patch method. These results suggest that the risk of aneurysm formation may be greater when large blocks of aortic wall are reimplanted.
Success has been reported with segmental artery reimplantation at specific levels to prevent postoperative paraplegia [10, 12]. However, the critical artery involved in protecting against spinal cord ischemia is difficult to identify. As noted, use of neurophysiologic tools to measure SEPs and MEPs is one valuable method [17–20]. Another method involves the injection of hydrogen into the clamped aortic segments and detection of this tracer by an intrathecal platinum electrode [21]. In addition, preoperative selective spinal angiography has been used for precise identification of the blood supply to the spinal cord [22]. Given that the critical artery for spinal cord ischemia cannot always be accurately identified at present, many segmental arteries should be reimplanted because they may function as collaterals to the anterior spinal artery even if the critical artery is not included among them.
Selective perfusion of visceral vessels may play an important role in preserving organ function [23]. Nonetheless, we have in our experience thus far been unable to demonstrate significant differences in blood urea nitrogen, creatinine, glutamic-pyruvic transaminase, and bilirubin in patients receiving perfusion versus those who did not. However, the technique of visceral perfusion does enable surgeons to perform segmental artery reimplantations in a more deliberate manner, without so much concern that visceral ischemia may develop.
As there is some risk of thromboembolism on applying the aortic clamp forceps to the atheromatous aortic wall, we have tried to apply the aortic clamp forceps to avoid the aortic sites with intraluminal thick atheroma, as detected by preoperative enhanced computed tomography. However, there were no cases in our study involving patients who had thick atheroma over the whole extent of the aneurysm, and the clamp forceps could be applied to aortic sites with relatively thin thrombus. The possibility of thromboembolism arose in our 1 patient who died of peritonitis (case 36). This patient underwent emergency laparotomy after surgery, and was found to have a pinhole perforation in the serosal surface of the small intestine. However, the excised short length of small intestine containing the perforation showed no evidence suggesting the presence of visceral arterial thromboembolism.
Limitations of this paper
Because this study includes only 51 cases, it cannot be stated definitively that our operative method will be uniformly successful in preventing postoperative paraplegia during TAAA surgery. However, it is noteworthy that we have not experienced any neurologic abnormalities after surgery in any of the patients.
Conclusions
In light of these clinical results, reimplantation of multiple intercostal or lumbar arteries using the sequential multisegmental aortic clamping method under distal aortic perfusion appears to be useful for preventing spinal cord ischemia.
Footnotes
1 This article has been selected for the open discussion forum on the CTSNet Web site: http://www.ctsnet.org/discuss 
References
This article has been cited by other articles:
|
|
 |
|
  R. K. Greenberg, Q. Lu, E. E. Roselli, L. G. Svensson, M. C. Moon, A. V. Hernandez, J. Dowdall, M. Cury, C. Francis, K. Pfaff, et al. Contemporary Analysis of Descending Thoracic and Thoracoabdominal Aneurysm Repair: A Comparison of Endovascular and Open Techniques Circulation, August 19, 2008; 118(8): 808 - 817. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Kawamura, H. Ogino, H. Sasaki, H. Matsuda, K. Minatoya, H. Tanaka, and S. Kitamura Spinal cord malperfusion caused by using the segmental clamp technique during descending aortic repair for chronic type B aortic dissection Interactive CardioVascular and Thoracic Surgery, February 1, 2008; 7(1): 146 - 148. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. James, M. Kuduvalli, J. Y Lu, and A. Rashid Coincidence of Spinal Canal Stenosis and Thoracoabdominal Aortic Aneurysm Asian Cardiovasc Thorac Ann, June 1, 2007; 15(3): 255 - 257. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Ogino, H. Sasaki, K. Minatoya, H. Matsuda, N. Yamada, and S. Kitamura Combined Use of Adamkiewicz Artery Demonstration and Motor-Evoked Potentials in Descending and Thoracoabdominal Repair Ann. Thorac. Surg., August 1, 2006; 82(2): 592 - 596. [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 |
