| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 1997;64:1555-1558
© 1997 The Society of Thoracic Surgeons
Department of Surgery, University of Virginia School of Medicine, Charlottesville, Virginia
Abstract
Thoracic aortic surgery has evolved from a high-risk, high-morbidity procedure to a safe procedure with predictable results. The frontiers left are adequate spinal cord protection and a less invasive approach to these procedures.
I want to start this discussion by thanking Dr W. H. Muller, Jr, for all that he contributed to my surgical career. He was a spectacular and innovative surgeon. He made many contributions to thoracic aortic disease that will be listed in the following section. One of my first cases at the University of Virginia was an abdominal aneurysm repair with Dr Muller and I was very impressed with the dexterity and smoothness with which he accomplished this task. Doctor Muller's surgical technique came from the days when one had to get in and get out in a hurry to accomplish the surgical goals. That is probably still an excellent philosophy, and he taught me this extraordinarily well. I had the opportunity to scrub on the last few open heart procedures that Dr Muller did at the University of Virginia. I always will feel honored for having had that opportunity.
The following discussion will be broken down into aspects of surgery for chronic thoracic aortic aneurysms. There will be separate sections on the descending thoracic aorta (including thoracoabdominal aneurysms) and another section on the ascending aorta. The discussion will start with a historical perspective and will evolve into newer techniques used. This synopsis will emphasize aspects of both spinal cord and cerebral protection during thoracic aortic operations. There will be no discussion in this section on acute dissections and other acute injuries of the thoracic aorta.
Descending Thoracic Aortic Aneurysms: Development of the Present Surgical Approach
One of the major hindrances to initial surgical approaches to the thoracic aorta was concern about paraplegia. Much of the experimental surgery was done on dog models, and a dog's spinal cord is extremely sensitive to ischemia. Only 10 minutes of ischemia results in paraplegia in dogs. It was not obvious to many of the pioneers in this area that the human spinal cord might be more tolerant of ischemia than noted in canine experiments. Therefore, the major developments initially were the concepts of working inside abdominal aneurysms. Matas in 1902 [1] and Creech [2] popularized this technique in aortic surgery. The concept of not resecting the whole aneurysm and dealing with intercostal vessels from within is absolutely critical to modern aortic surgery. This is in contrast to early attempts at wrapping descending thoracic aneurysms with, unfortunately, inconsistent results [3]. Direct resection was popularized by several pioneers including DeBakey and Cooley [4], who used aortic allografts to replace the thoracic aorta. Bahnson [5] used a technique of lateral resection and direct closure.
The thoracoabdominal aorta provided even more challenges. Many of the initial operations related to separate grafting of all of the visceral vessels [6]. Crawford and colleagues [7] developed a modern technique of working within the aneurysm and used the inclusion technique to graft the visceral vessels within the aorta.
Spinal Cord Protection
Clearly, the major obstacle to successful operation on the thoracic aorta still is the prevention of paraplegia. There are multiple series that have discussed the risk of paraplegia during thoracic aortic operations, but the rate of this complication varies in most series between 5% and 15%. Crawford and colleagues demonstrated a useful classification for thoracic and thoracoabdominal aneurysms: type I, proximal descending thoracic aorta to the upper abdominal aorta; type II, proximal descending thoracic aorta to below the renal arteries; type III, distal half of descending thoracic aorta into the abdomen; and type IV, most or all of the abdominal aorta. The usefulness of this classification relates to the risk of paraplegia with resection using the clamp and sew techniques. The rates of spinal cord injury for type I through type IV were 15%, 31%, 7%, and 4%, respectively [8]. The mechanisms of spinal cord injury are still controversial. The blood supply to the spinal cord is inconsistent in humans. The superior cord (C1-T3) is well supplied with three to five anterior radicular arteries. The midthoracic cord (T4-T8) has, at most, one anterior radicular artery. The remaining lower thoracic and lumbar spinal cords are supplied with three to five anterior radicular arteries, including the most important one, which is the artery of Adamkiewicz. This artery originates in most humans between T8 and L3 [9]. Although this artery can be identified angiographically, Griepp and colleagues [10] have suggested that it may not be physiologically important.
However, it is clear that ischemia does lead to spinal cord injury. It has been noted that cross-clamp times of less than 30 minutes have a much less risk of paraplegia than those with clamp times above 30 minutes. Svensson and colleagues [8] noted that cross-clamp times less than 30 minutes resulted in less than 10% risk, whereas those greater than 60 minutes had more than a 20% risk. Livesay and associates [11] noted that the risk of injury increased from 3% to 11% when a cross-clamp time exceeded 30 minutes. In our series of patients undergoing thoracic aneurysm repair using the clamp and sew technique, my colleagues and I noted no definite correlation between cross-clamp time and neurologic injury. However, we had short aortic clamp times at a mean of about 30 minutes for all patients. We did note spinal cord injury rate of 9% in the study group. We noted injury occurred even in clamp time as short as 20 minutes [12].
Clearly, injury can occur from mechanisms other than pure ischemia. Reperfusion injury has been noted as a possible mechanism for spinal cord injury. We have noted in experimental work that hyperemia after spinal cord ischemia correlated with neurologic deficits. We suggested that there may be a development of spinal cord edema and a possible compartment syndrome in those animals in which spinal cord injury developed [13]. Blaisdell and Cooley [14] have suggested spinal cord drainage to prevent this complication. Svensson and colleagues [15] demonstrated that this technique did not result in improvement in spinal cord outcome.
Mechanical Protection of the Spinal Cord
Many individuals have gone to mechanical methods of maintaining spinal cord blood flow. The Gott shunt is a shunt that flows from either the left ventricular apex or the aorta above the clamp to the femoral artery or descending aorta below the clamp. Although Verdant and associates [16] confirmed excellent results with the Gott shunt, many surgeons use the technique of left atrial–femoral bypass using a centrifugal pump. This provides excellent hemodynamics and reduces left ventricular hypertension after placement of the aortic cross-clamp. Although Crawford and colleagues have not demonstrated any obvious improvement in results with this technique, it is probable that most surgeons in the country use some mechanical support like left atrial–femoral bypass during thoracic aortic operations. Finally, others have advocated use of femoral–femoral bypass to provide some flow below the cross-clamp. This technique probably is most helpful in those situations where the aorta cannot be clamped proximally and circulatory arrest is required to do the procedure. Therefore, the patient can be cooled with femoral–femoral bypass and, eventually, the procedure can be done after the pump is turned off. This does lead to the risk of embolizing debris retrograde from the aneurysm with the potential of this going to the brain. We have seen this complication, and it is a real risk of femoral bypass in individuals with chronic thoracic aneurysms.
Pharmacologic Protection of the Spinal Cord
Many have considered the use of pharmacologic manipulations to reduce spinal cord ischemia. We and others have been very interested in methods of increasing spinal cord tolerance to ischemia. Many surgeons have used hypothermia, and we have experimentally tested several techniques including barbiturates as well as free radical inhibitors [17]. Each has had reasonable results in experimental models, but none were completely protective. Most recently, we developed the concept of using a cold adenosine solution as a method of improving tolerance of the spinal cord to ischemia [18]. This is based on the experimental efficacy of adenosine in myocardial protection. We were amazed to note that adenosine was completely protective in a very sensitive rabbit model of spinal cord ischemia. The animals were given hypothermic adenosine solution over the first 10 minutes of ischemia. This was contrasted to one control group of animals that had the aorta cross-clamped without infusion of adenosine or saline solution. Another control group had the aortic segment flushed with normothermic saline solution. All of the animals receiving the adenosine solution were completely free of spinal cord injury, whereas control groups were completely paraplegic. We have not attempted to use this technique clinically thus far, but we believe this may be a promising technique for spinal cord protection in humans.
Modern Surgery for Aneurysms of the Descending Aorta
Many techniques can be used successfully. Our preference presently is the clamp and sew technique. The patient is prepared with a double-lumen endotracheal tube, and monitoring of both pulmonary artery pressures and arterial pressures is performed. Before placement of the cross-clamp, the blood pressure is brought down systemically with nitroprusside and then the clamps are applied. The operation is accomplished as rapidly as possible and clamps are then removed. We make every attempt to reimplant intercostal vessels either at the proximal or distal suture lines, although we rarely will reimplant a very large intercostal vessel in the middle of a repair. This technique has led to a mortality of approximately 10% and a spinal cord injury rate of approximately 9%. The spinal cord injury rate is less than 7% in the survivors, with most of injuries not being complete paraplegia. It is clearly acceptable to use left atrial–femoral bypass, Gott shunts, or peripheral cardiopulmonary bypass for these repairs. However, one should certainly avoid clamping the aneurysmal thoracic aorta while doing proximal anastomosis for fear of releasing debris to the visceral vessels as well as the spinal cord blood supply.
Surgery of the Ascending Aorta
Historical Perspectives
Cooley and DeBakey [19] and Bahnson [20] separately reported lateral resection of aneurysms of the ascending aorta. Muller and associates [21] described the technique of ascending aortic resection and bicuspidization of the aortic valve to treat aortic insufficiency in patients with Marfan's syndrome. In this initial series, they described 2 cases, 1 ascending aortic replacement with a woven Teflon graft and 1 treated with aortorrhaphy. Bentall and de Bono [22] first described treating annuloaortic ectasia and aortic regurgitation with a composite valve and tube graft. Muller and associates [23] described successful replacement of the aortic arch with several side grafts without the use of cardiopulmonary bypass. This was done by bypass of the involved area and finally aneurysm exclusion. In 1 of the cases, Muller and associates attempted to use hypothermia, but this was unsuccessful. Finally, DeBakey and Cooley [24] reported successful aortic arch resection using cardiopulmonary bypass and an allograft to replace the arch.
Modern Surgical Approach to the Ascending Aorta
There have been several modifications of techniques that have allowed for extremely successful surgery of the ascending aortic aneurysms. Clearly, aneurysms without valvular involvement are very simple to deal with presently. Typically, the patient is cooled to 28° to 30°C. Aprotinin may be given to allow for better hemostasis. The use of albumin-impregnated grafts reduces bleeding complications. Finally, in most situations, the aorta can be cross-clamped and the aneurysm can be resected and replaced with a graft with minimal morbidity and mortality.
When the aortic root is involved with aneurysm disease, things get more complex. Certainly, composite graft resection has been well described. Our technique presently is the use of the composite graft with separate implantation of both coronary ostia, which have been dissected free from the aortic wall. The aorta is transected distally for a much simpler anastomosis. We have used a second-layer closure on the proximal site to avoid bleeding in this difficult to access area [25]. Previously, the technique was that of an inclusion type where the coronary ostia were put directly in without being resected from the aortic wall. This allowed for quite good hemostasis, but Kouchoukos and colleagues [26] demonstrated that there was an unacceptably high incidence of false aneurysms. This, therefore, pushed them to recommend the technique that we use today. In our own present series of resections of the ascending aorta for aneurysm disease, we have operated on 85 consecutive patients with only one death. We believe, therefore, that this is an extremely safe operation even in patients who could otherwise be quite high risk.
Surgery of the arch clearly still is more complicated. Circulatory arrest has been advocated by Ergin and coworkers [27], and they have described the technique of cooling and ice around the head for the best overall results. Clearly, circulatory arrest has made this operation a great deal easier. Ott and associates [28] have described the hemiarch technique in which a single anastomosis is made by careful tailoring of the arch and removal of all aneurysmal tissue. However, circulatory arrest is still needed in this situation.
A major advance in the treatment of the arch aneurysms requiring circulatory arrest is the use of retrograde cerebral perfusion. This technique is performed by circulating aterial blood slowly through the superior vena caval cannula after arterial flow has been stopped. The flow is maintained usually at a rate of 200 to 400 mL per minute and the central venous pressure is maintained at less than 30 cm H2O. It is unclear if this technique actually perfuses the brain. However, it most likely helps deair and prevents debris from flushing into the cerebral vasculature. Our perception is that patients do extremely well with this technique and seem to wake up sooner than they would have if straight circulatory arrest had been used.
The final modification to be discussed is the so-called elephant trunk technique described by Borst and associates [29]. This is used in the situation where the patient may have an associated aneurysm of the descending thoracic aorta but the arch and ascending aorta need to be replaced. A small section of graft is left distally beyond the most distal anastomosis. This makes it much easier when the descending thoracic aneurysm is repaired at a later date to avoid a great deal of dissection. The descending aorta is quickly opened and the section of graft is grasped and clamped. This, therefore, will simplify a later operation for patients with very diffuse aneurysmal disease.
Summary
In summary, surgery for thoracic aortic disease has progressed rapidly due to early pioneering efforts. The last frontiers clearly are better methods of spinal cord protection and less invasive techniques of thoracic aortic surgery.
Footnotes
Presented at Cardiovascular Surgery—Then and Now, University of Virginia Medical Center, Charlottesville, VA, April 26, 1997.
Address reprint requests to Dr Kron, Department of Surgery, University of Virginia Medical Center, Box 310, Charlottesville, VA 22908.
References
This article has been cited by other articles:
|
|
 |
|
  D. Kalavrouziotis, R. J.F. Baskett, and J. A.P. Sullivan Pulmonary artery to distal bypass for surgery on the descending thoracic aorta Interactive CardioVascular and Thoracic Surgery, June 1, 2005; 4(3): 170 - 172. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. A. Hessel Bypass Techniques for Descending Thoracic Aortic Surgery Seminars in Cardiothoracic and Vascular Anesthesia, November 1, 2001; 5(4): 293 - 320. [Abstract] [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 |
