| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||
Ann Thorac Surg 1997;64:1675-1677
© 1997 The Society of Thoracic Surgeons
DR RANDALL B. GRIEPP (New York, NY): I congratulate Dr Deeb and his colleagues for bringing to our attention the difficulties in managing a very high risk group of patients: those with acute type A dissection and malperfusion. The authors' conclusion that delayed operation is preferable to immediate surgical intervention is a provocative one and must be critically examined, because it represents a significant change from the policy in most surgical units.
By way of comparison, I examined our experience with acute type A dissection over the past 10 years. All of our patients were operated on as soon as possible, with the diagnosis being made primarily on the basis of clinical examination findings and confirmed by transesophageal echocardiography. Of 120 patients, 18 died, for a 15% surgical mortality. Thirty-seven patients had malperfusion; the mortality in them was 27%, constituting over half the deaths occurring overall. In the group without malperfusion, the mortality was 10%. Our experience differs from that of Dr Deeb and his colleagues of Ann Arbor: Although our immediate surgical treatment of patients with malperfusion resulted in a 27% mortality, which is not dissimilar to his statistic for the delayed approach to malperfusion, it differs markedly from the mortality of 89% in the malperfusion patients Dr Deeb treated with early operation.
To try to explain the differences in our results, I would like to direct a number of questions to Dr Deeb. In the immediate surgical group with the very high mortality rate, was there some delay in getting patients to the operating room occasioned by difficulties in transferring them from another institution, with diagnostic procedures, and so on, so that the consequences of malperfusion would have become well established by the time of operation?
Was the extent of the operation in these patients appropriate for such compromised patients? Would a quicker, less extensive, albeit admittedly less durable, repair have minimized operative potentiation of malperfusion injury?
Was the operative procedure designed to promptly reverse malperfusion with cardiopulmonary bypass? If perfusion was instituted from the femoral artery, was transesophageal echocardiographic monitoring of perfusion patterns in the descending thoracic aorta carried out? In patients in whom satisfactory perfusion by femoral cannulation could not be accomplished, was the cannula switched into the ascending aorta or some other maneuver used to minimize the intraoperative impact of malperfusion?
Was malperfusion prevented during rewarming by placing the cannula in the ascending aortic graft, thereby minimizing malperfusion at that time?
Why was the incidence of coronary artery bypass grafting so high in these groups? It was used in 5 of 9 of the patients in the immediate repair group, but only 2 of these patients had myocardial malperfusion preoperatively. In the delay group, 7 of 17 had coronary artery bypass grafting, and again only 3 of these had myocardial malperfusion preoperatively.
Do you believe that the percutaneous treatment of malperfusion can be recommended? From the manuscript, which Dr Deeb very kindly provided to me, 11 of the 20 patients in the delay group had fenestration and 5 of these had stent placement. This group, however, included all of the deaths, five, yielding a 46% mortality in the patients treated percutaneously. No deaths occurred in the subgroup of patients with delayed operation who did not undergo fenestration or stenting, so I think the contribution of fenestration and stenting needs to be defended.
Finally, I would like to acknowledge Dr Deeb and associates' many important contributions to the management of acute dissection and to applaud their iconoclasm, without which progress in this field is not possible.
DR D. CRAIG MILLER (Stanford, CA): I also congratulate Dr Deeb and his colleagues for bringing this interesting and provocative information to our attention.
Number one, they have reemphasized what we have known for a long time, that acute dissection when coupled with these devastating malperfusion complications is truly a very lethal disease. This combination carries, based on our historical data and those of others, a predicted mortality rate of around 50%, but Dr Deeb, not in the 80% or 90% range, as in your historical surgical comparison group at Michigan. Unfortunately, the lawyers in North America also know well the natural history of acute type A dissection; would you have had any reasonable legal defense if one of your patients had succumbed to an ascending aortic rupture while you were waiting for the distal malperfusion problems to resolve?
Number two, and perhaps most importantly, I commend you and your interventional radiology colleagues for demonstrating how percutaneous endovascular flap fenestration, with or without stenting of either the true or false lumen, or both, has been such a tremendous recent advance in this field. Indeed, your radiology co-author, David Williams, is one of the cardiovascular interventional radiology pioneers. It is very quick, it is very effective, and it is truly life-saving.
At Stanford now, the cardiovascular surgical service has referred over 50 patients to our cardiovascular interventional radiology colleagues for dissection flap fenestration or stenting, or both, over the past 3 years. Most of these patients had acute dissections. Almost all of the acute type A patients had this done after the ascending aorta had been replaced, in contrast to Dr Deeb's approach. In many cases it was within hours of the ascending aortic surgical procedure. Conversely, in patients with acute type B dissections, none had undergone any previous thoracic surgical procedures. What do you think of performing "fen/sten" very early postoperatively?
Finally, I am afraid I cannot agree with your conclusion that we should delay aortic replacement for these people, even though we realize they are very high risk surgical candidates. In your statistical analysis I think you may have "stacked the deck," in that, as I interpret your numbers, only two of the eight deaths in your earlier surgical experience can truly be blamed on distal malperfusion complications. It is hard for me to understand how low cardiac output and failure to wean from cardiopulmonary bypass can be blamed on distal thoracoabdominal malperfusion. In a way, it is analogous to saying you are going to watch all postinfarct ventricular septal defects for 6 weeks. Sure, the operative results may look great, but your overall salvage rate is relatively dismal.
Again, Dr Deeb, thanks for introducing this iconoclastic approach to a very difficult and challenging clinical problem.
DR DEEB: I thank Dr Griepp for his comments. Doctor Griepp has asked many interesting and intriguing questions, and I will try to answer them in a concise and orderly fashion. Before answering his questions, I would like to congratulate Dr Griepp and his group on the outstanding results in their surgical treatment of patients with acute type A dissection and malperfusion. However, before I compare his excellent results with those from the University of Michigan, I would like to be certain that we are comparing two similar patient populations. Clearly we are not including patients with malperfusion and no end-stage organ failure. The patient who presents with an acute type A dissection and who has signs and physical findings consistent with malperfusion, but not end-stage organ failure, would not qualify for surgical delay. We believe that this patient should undergo immediate surgical repair of the acute type A dissection and would be expected to fall within the 25% mortality range. However, patients presenting with signs and physical findings consistent with malperfusion in association with end-stage organ failure are included in the population that we are proposing for surgical delay. Clearly, what we are discussing today are patients who present with an acute type A aortic dissection with malperfusion and a dead or near-dead heart, limb, or visceral organ. In addition, we are including patients who have a severe neurologic deficit and not patients with transient neurologic changes. I therefore request that Dr Griepp reexamine the patient population he has described to be certain that his patients fall in the same categories as our patients. Clearly if his patients meet the same criteria as our patients and he still maintains a 27% mortality rate with immediate operation, then I believe our group has a lot to learn from Dr Griepp and his colleagues in New York. However, if he too finds that he has a significant mortality of 75% or more in the patients who present with malperfusion and meet the same criteria as those in our group, then I believe he ought to consider percutaneous reperfusion and delay of repair.
In response to the first of Dr Griepp's many questions, it is true that we are a tertiary-care referral center for the state of Michigan. Most patients with acute type A dissection and malperfusion referred to the University of Michigan arrive at the University of Michigan after a considerable delay. The patients with limb ischemia came between 36 to 48 hours after the acute insult. At the time of presentation these patients had severely compromised limbs and they were taken immediately to the operating room, where surgical repair of their acute type A aortic dissection was performed as well as revascularization of their limbs. This same criteria held for the other patients in the non-delay group, who had either severe myocardial compromise resulting from acute myocardial infarction or significant abdominal ischemia with markedly compromised bowel or visceral organs. These patients also were taken to the operating room for surgical repair of their type A aortic dissection, with revascularization of the above-mentioned organs and with the same poor end result. It became quite evident to us that patients with severely compromised malperfused organs who underwent immediate surgical repair of the aortic dissection did not fare well, and that an alternative approach to these patients may be in order.
In response to Dr Griepp's second question, we did try to do the quickest operation in each patient, and we thought that a quicker, less extensive procedure might be more advantageous than a more durable long-term repair. Our statistical analysis showed that the patients in the delayed group had a significantly different and more complex operation than those patients who underwent immediate repair during their acute crisis with malperfused organs.
With regard to Dr Griepp's third question, our operative procedures were designed to promptly reverse malperfusion with cardiopulmonary bypass. All patients were placed on cardiopulmonary bypass using femoral arterial access, and echocardiography was used in 100% of these patients. We were confident we obtained intraoperative perfusion of ischemic organs with the onset of cardiopulmonary bypass.
In response to Dr Griepp's fourth inquiry, it is standard policy at the University of Michigan to switch the arterial cannula into the graft immediately after the distal anastomosis is performed and systemic warming is begun. Not only are we assured of a higher probability of organ perfusion, but patients also warm more quickly, thereby decreasing the operative time.
Regarding Dr Griepp's fifth question, the incidence of coronary artery bypass grafting was high in the non-delay group because of the circumstances at the time of the dissection. Three of the 9 patients had myocardial malperfusion and required coronary artery revascularization. In the other 2 patients the dissection extended to the right main coronary artery, and during implantation of the right coronary button into the composite graft, it was decided that the tissue was inadequate and a right coronary artery bypass graft was performed.
Finally, to answer Dr Griepp's last question concerning the justification for percutaneous fenestration and reperfusion, because all the patients in the delayed group who died underwent this procedure, I can only say that we believe expeditious reperfusion of the malperfused organs is essential for survival. Eleven of the 20 patients did undergo fenestration and stenting, and the other 9 either spontaneously reentered before fenestration or had myocardial malperfusion with a completed myocardial infarction or neurologic malperfusion with stroke, in which case reperfusion was considered inappropriate. Clearly all the patients with limb and visceral malperfusion underwent reperfusion, and these patients did not die secondary to the reperfusion procedure. Three of the patients died preoperatively, 1 of rupture and tamponade and 2 others of multiorgan failure secondary to ischemic reperfusion injury. The 2 patients who died of a reperfusion injury with percutaneous reperfusion would have suffered the same fate had they undergone immediate surgical repair and reperfusion, at a considerable increase in the medical resources utilized. The 2 other patients whose condition stabilized after percutaneous reperfusion resolved all of their reperfusion injury died at the time of their surgical procedures as a result of complications not associated with fenestration and reperfusion.
In response to Dr Miller's question regarding the option of surgically repairing the acute type A dissection, followed by postoperative fenestration, this technique was employed in 2 patients in the non-delayed group who died. Both of these patients had limb malperfusion and underwent immediate repair of their type A aortic dissection. One patient underwent femoral-femoral bypass at the time of operation. After repair, they both continued to have malperfusion of their limbs and underwent fenestration. Unfortunately, both of these patients died of severe multiorgan failure secondary to ischemia reperfusion injury.
With regard to Dr Miller's question about the fact that patients died of cardiogenic shock and how this could be related to distal organ malperfusion, I would like to elaborate on the mechanisms of ischemia-reperfusion injury. In patients who had limb and visceral malperfusion and who underwent successful reperfusion, a significant pulmonary capillary leak developed, perhaps secondary to the high levels of circulating cytokines and adhesion molecules produced in association with the up-regulation and activation of complement. There are many animal models in which the mechanism of end-stage pulmonary failure with ischemia-reperfusion of visceral and limb organs can be substantiated. Not only does fulminant pulmonary edema develop in patients as a result of the capillary leak syndrome, but considerable pulmonary hypertension with a marked elevation in the pulmonary vascular resistance develops as well. Because the right heart is purely a volume ventricle and not a pressure ventricle, it cannot withstand the acute onset of increased pulmonary vascular resistance, and subsequently these patients go into low-output and cardiogenic shock resulting from acute right ventricular failure. Also, the high levels of circulating cytokines occurring secondary to the acute inflammatory response of ischemia reperfusion injury may be direct myocardial depressants and, in association with the elevated pulmonary vascular resistance, augment acute right ventricular failure.
In those patients who suffered malperfusion of their hearts, myocardial reperfusion caused direct ischemia-reperfusion injury to the heart analogous to the injury noted in patients who undergo acute revascularization after myocardial infarction. This, in association with elevated pulmonary vascular resistance and capillary leak syndrome, would lead to cardiogenic shock and low output. Therefore the cause of death in laboratory models of ischemia-reperfusion injury as well as in human patients suffering from ischemia-reperfusion injury is acute pulmonary failure with the capillary leak syndrome and elevated pulmonary pressures and resistance, which lead to the development of severe acute right ventricular failure and cardiogenic shock.
In conclusion, I thank both Drs Griepp and Miller for their insightful comments and excellent questions concerning this very problematic situation and highly controversial topic. I also thank The Society for the opportunity and privilege of presenting our paper.
Related Article
Ann. Thorac. Surg. 1997 64: 1669-1675.
| ||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
