Portal de Periódicos da CAPES

Panel Discussion: Session II—Aortic Arch

2007; Elsevier BV; Volume: 83; Issue: 2 Linguagem: Inglês

10.1016/j.athoracsur.2006.11.093

1552-6259

Randall B. Griepp, Robert S. Bonser, Axel Haverich, Teruhisa Kazui, Nicholas T. Kouchoukos, Hazim J. Safi, Hiroaki Sasaki, David Spielvogel, Shinichi Takamoto,

Cardiac Valve Diseases and Treatments

DR RANDALL B. GRIEPP (New York, NY): As the essayists are assembling, let me just ask a few questions, for my own information, of the audience with regard to arch surgery. Could I just have a show of hands of those of you who do a reasonable number of arch replacements or feel comfortable doing arch surgery? It looks like perhaps half. Let me ask then, of those of you who responded, how many of you rely just on hypothermic circulatory arrest and do not use any other type of cerebral protection: HCA only? A pretty small number, maybe 5%. How many of you use retrograde cerebral perfusion at least in some of your patients or at some point in the operative procedure? It looks like 15% or 20%. How about antegrade cerebral perfusion? That looks like probably the other 80%.Let me just ask one multipart question of the panelists, if I could, just as sort of a poll of our experts. Would any of you use any form of cerebral protection other than HCA if you were confident that your period of cerebral ischemia is going to be limited to less than 30 minutes? That is the first part. Secondly, those of you who use selective cerebral perfusion, what do you use as the parameters for the perfusion: how do decide on flow, pressure, temperature, etc? What are the parameters that you use? And finally, do you use any form of monitoring to assure the adequacy of the perfusion that you are providing? This is with selective cerebral perfusion.DR NICHOLAS T. KOUCHOUKOS (St. Louis, MO): As far as less than 30 minutes for hemiarch and procedures that do not require an extensive replacement of the entire arch, we are comfortable with hypothermic circulatory arrest and use retrograde cerebral perfusion only at the end of the ischemic interval to evacuate air and debris. In terms of the parameters of perfusion when we use antegrade cerebral perfusion, we will monitor the arterial pressure from the opposite radial artery. In other words, if we are using the right axillary artery, we would monitor the left radial artery. And also we use Dr Kazui’s formula of approximately 10 cc/kg/min of flow during that interval, maintaining it at 20°C. And for monitoring, I think we showed you that we used the Somanetics system.DR SHINICHI TAKAMOTO (Tokyo, Japan): As I showed in the slides, circulatory arrest is inferior to the other protective procedures. So that even for less than 30 minutes, continuous retrograde cerebral perfusion is selected. With regard to the parameters for retrograde cerebral perfusion, I use the central venous pressure as a good index. I use Hamamatsu Photonics near-infrared monitors. It is a better system than the Somanetics because we can obtain the actual saturation of the tissues.DR HAZIM J. SAFI (Houston, TX): Well, I am the oddball here: we use retrograde whether it is 30 or 70 minutes. But we discovered that you must monitor brain flow using a transcranial Doppler, and this is whether you use the axillary or the ascending or the femoral artery. My associate, Dr Tony Estrera, did a study. About 58 patients over 2 years were divided equally among retrograde and antegrade perfusion, and there was no difference between them. But what we found was that even with 10 cc/kg/hr—we use the Kazui formula—there may be no flow during antegrade perfusion. You have to be sure there is flow in the middle cerebral artery, and you have to use a transcranial Doppler. We cannot compare apples to oranges. We use all these monitoring methods, including transcranial Doppler, and INVOS. We have information on about 200 patients. For 30 minutes I will use retrograde because the incidence of stroke is so low. With antegrade, we like to monitor using the transcranial Doppler, and with somebody who understands it, not anesthesiologists who tell you if there is a flow or not, but somebody who is a technician, who is trained. And we find a lot of pitfalls, even when we rewarm. Sometimes the flow goes to the subclavian artery, sometimes it goes to the right, but doesn’t go to the left. We found that in 20% to 25% of our patients, we changed the method of perfusion in response to monitoring.DR AXEL HAVERICH (Hannover, Germany): In 1999, Jean Bachet, after we presented our results in aortic arch surgery with a nearly 20% neurological complication rate (not all of them permanent, of course), said you have to change and do antegrade perfusion, which we did. We are a teaching institution, and we train a number of residents and also foreign and domestic fellows. So we said if we need to do circulatory arrest, we will use antegrade cerebral perfusion. This may or may not be correct, but we do it even if we have a circulatory arrest period of less than 30 minutes. We use 10 cc/kg/min, monitoring the radial artery. We also use oxygen saturation in both hemispheres with the Somanetics system for monitoring.MR ROBERT BONSER (Birmingham, United Kingdom): We would only use circulatory arrest if it was for a very short period. We are always prepared to have the SACP circuit there, so HCA alone is really only used when there is an open distal anastomosis to be performed. We use 10 ml/kg/min, just as Professor Kazui has stated. We have both corporeal and perfusate temperatures at 15°C, with a hematocrit of 20–30%, and we monitor the pressures. We would like to also monitor the middle cerebral artery velocity bilaterally, and we are training our anesthesiologists to be able to do that.DR DAVID SPIELVOGEL (Valhalla, NY): To answer the first question, for less than 30 minutes of circulatory arrest, we use no additional perfusion techniques. The head is packed in ice. I think it is important to prevent rewarming. We do cannulate the axillary artery in almost all aneurysm surgery, whether it is a hemiarch or a full arch, and I think that is important for preventing embolization during the open anastomosis.In terms of selective cerebral perfusion, we use perfusate at approximately 8 to 10 cc/kg/min, giving you a mean pressure anywhere in the 40 to 60 range, depending on the patient. The perfusate temperature is about 18° to 20°C. I think it is important not to overperfuse the patient during this period because that can cause neurologic sequelae as well.DR TERUHISA KAZUI (Hamamatsu, Japan): We use DHCA without RCP in selective cases, as I said in my presentation, in which we expect cerebral protection time of less than 25 or 30 minutes. As far as SCP is concerned, on the basis of our experimental study, we perfuse the bilateral two-arch vessel and the innominate artery, or the right axillary artery and left common carotid artery, at a flow rate of 10 cc/kg/min at a rectal temperature of 25°C. And right radial output artery pressure as well as the bilateral catheter tip pressure are adjusted to about 40 mm Hg. If the patient has a history of old cerebral stroke and hypertension, the safe lower limit of autoregulation shifts to the right, so we increase the perfusion pressure to about 50 to 70 mm Hg by increasing our perfusion flow. And monitoring of SCP includes the right radial perfusion pressure as well as the bilateral catheter tip pressure and flow rate, and we have serial electroencephalography and near-infrared spectroscopy. If available, we monitor SjO2 entering, and transcranial Doppler sonography to check the flow velocity in the middle cerebral arteries.DR HIROAKI SASAKI (Osaka, Japan): We perform deep hypothermia with selective cerebral perfusion. The temperature previously was at 20°C, but recently we increased it to 28°C, but we added left subclavian artery perfusion, and the perfusion pressure has been gradually increased. The perfusion flow is 15 cc/kg/min.DR GRIEPP: Just in commenting, Dr Miller pointed out this morning that Dr Kazui’s patients and some of ours are different sizes. Our average patient weighs over 80 kilos, and most of that isn’t in their heads. So we tend to rely on the pressure rather than the flow, but I think your perfusionist should be aware of both. There are pitfalls both ways.DR CATHARINA A. VAN DOORN (London, United Kingdom): This is a question to the panel in general on the prevention of secondary brain injury. After deep hypothermic circulatory arrest, reperfusion can cause secondary brain injury; in particular, the presence of hyperthermia is a risk factor for this. Following deep hypothermic circulatory arrest, we would therefore now generally rewarm the patient to only 36°, and if there has been a long period of arrest, then the patient will be kept at 36° for 24 hours in the intensive care unit. What is the view of the panel on the prevention of secondary brain injury and what methods do they use?DR SAFI: I think the best thing is to prevent it, and the way you prevent it is to monitor. I am still in love with this transcranial Doppler because our Hungarian doctor who monitors it can tell us whether there is cerebral edema or not, and it really makes a difference. And we have our neurologist attached to our patients to check on them.DR SPIELVOGEL: There are a few things I think that are important. When you come off a period of circulatory arrest, it is helpful to perfuse perhaps for a few minute with cold perfusion before rewarming if you have completed your repair. It is also helpful not to rewarm the patient all the way. We tend to come off bypass when the tympanic temperature is about 32.5°C and the bladder temperature just over 35°C, and allow the patient to rewarm slowly in the ICU. We don’t really see any increase in bleeding associated with this, and I think it is better for the brain not to fully rewarm. And of course your perfusionists have to be careful about the temperature differential while rewarming.MR BONSER: The only point I would add to that is the importance of maintaining circulatory stability during that vulnerable period. The patient goes to the intensive care unit and has very frequent output monitoring. We do actively warm them with bear huggers, and we try to make sure there is adequate flow and pressure throughout that period until they’ve equalized their peripheral and central temperatures and they are ready to awaken.DR HAVERICH: I think one important asset we acquired with selective antegrade perfusion was the fact that we did not have to cool the patient down to 20 and 19 and 18°C any more. Now we cool the patient to 28°C. And the problem you are referring to is not so much of a clinical problem any more with higher temperatures during the period of circulatory arrest.DR SAFI: We accepted, and critically, that if you use antegrade perfusion, you don’t need to freeze the brain. But I think and still believe in what my mentor, Dr Crawford, called the cold fish principle. If you have a box in Houston, where the temperature is about 40°C, and you put a fish in it and then no ice, 4 hours later you can’t open the box. But if you stuff it with ice, I think you can eat that fish. So I think cold is good. I believe in it. What you add to it is secondary. I don’t think that using retrograde is going to protect the brain, because we don’t know how much flow goes to the brain. Most of the studies are not convincing.DR CHRISTOPHER PAUL YOUNG (London, United Kingdom): I have noticed several speakers who have advocated individual branch replacement of the aortic arch vessels. We have done about 210 intrathoracic stent grafts, and we have covered the left subclavian in 50 of those cases. We have had no sequelae. Does the panel really think that it is necessary to reimplant the left subclavian artery? Would it not be enough to oversew it and save time on the anastomoses?DR SAFI: It is really 5 to 10 minutes to sew the graft to the graft. Just keep it the way nature made it. It seems to me it is against surgical principle to see an artery to which you can put a graft and say, “No, I am going to tie it off.” That would be what I call the 1917 method of treatment. I don’t think 5 or 10 minutes is going to add much to the overall risk to the brain.DR SPIELVOGEL: I agree with Dr Safi. I think that you should maintain anatomy. I think it is important. The patient came to the operating room with a subclavian artery that was patent; it supplies the posterior circulation of the brain and the left arm. If you can reconstruct it in a timely fashion, I think that is optimal. If the patient, of course, has a patent internal mammary artery, I think it is probably a very good idea to maintain patency. When it comes to the question of stent grafts, it is true one can sometimes get away with it, but there are a certain number of patients who require subsequent subclavian carotid bypass. If you are going to do arch surgery, you don’t want to find out postoperatively—after you have tied off the left subclavian artery and compromised the posterior circulation—that a patient had some disease in their right vertebral artery of which you were unaware. So I agree you should maintain the anatomy. I think the techniques are such that you can do that in quite an expeditious fashion.DR GRIEPP: In addition to that, in patients with aneurysmal disease, you may have to deal with downstream aneurysms, and the subclavian arteries provide important collateral circulation to the spine. I would hesitate, in those patients that might require subsequent downstream reconstructions, to take out one subclavian artery if it is not really necessary.DR YOUNG: I would totally agree that when the disease process involves the low thoracic or thoracoabdominal region, that the incidence of paraplegia is liable to be raised. But I do wonder in view of the stent graft evidence, whether it is absolutely essential. I use the patch technique in arch reconstructions, and therefore it doesn’t really affect my practice.DR GILLES D. TOUATI (Amiens, France): Our experience concerning aortic arch reconstruction is modest. Twenty-six patients have been operated on, including complete aortic arch replacement with a Bentall or Tirone David procedure, and with the introduction of stenting, an uncovered stent in the descending aorta (like the Djumbodis system). All these patients were done without stress, and with no time limit because of a completely normothermic approach with antegrade cerebral perfusion (as we described in the Annals in 2003). We don’t know what really happens in autoregulation of cerebral blood flow during hypothermia. Our mortality and morbidity were zero. My question for all the surgeons on the panel, given that the latest guidelines for brain injury have abandoned hypothermia, whether they believe that the normothermic approach is the next logical step for a complete and more aggressive approach for aortic surgery?DR SAFI: Your approach is called retro. That is what Dr Cooley and Dr DeBakey did in the 1950s, with a high incidence of stroke and death. But you have very good results, so just stick with it.MR BONSER: It is fine as long as you don’t have a problem. But when you have a problem, the patient will have a problem. You have excellent results and you are to be congratulated for that, but one day some mishap or difficulty will occur, and then the patient will be in difficulty. So I cannot concur with your view.DR SAFI: There is a morbid fear of profound hypothermia. I have a letter that came to me from a famous artist. I did his ascending aorta and arch. We did retrograde, and went 57 minutes because we had to use branches and so on. He sent me a beautiful letter telling me his artwork has come flowing. He is 76, and he sent me a beautiful letter asking whether there might be a relationship between his productivity and the frozen brain. I told him, I am a secular man, so I don’t know what we did to these cells. But I would ask my associate Tony to freeze my brain when he does my ascending aorta and arch.DR GRIEPP: One other comment. When we do cognitive testing in some of our patients, even after we protect the brain in what we think is absolutely the optimal fashion, in some we can detect differences. There is no perfect way. We don’t have all the answers. Hypothermia does give you, in many cases, a little more margin for error, either surgical or judgmental.DR GIAMPIERO ESPOSITO (Lecce, Italy): A question for Professor Haverich. How long should the stent in the thoracic aorta in the frozen elephant trunk procedure be? Do you use any type of monitoring to see the deployment of the stent in the thoracic aorta?DR HAVERICH: We used fluoroscopy in the first 21 cases, and now do it with transesophageal echocardiography. We are very hesitant to deploy a stent longer than 12 cm because we are afraid of paraplegia.DR SAFI: That is a good point; you are right. We did an elephant trunk, in the same setting after coming off pump, stenting the whole descending aorta, and 24 hours later the patient developed delayed paraplegia.DR GRIEPP: Axel, I wasn’t quite clear during your presentation whether you anchor the elephant trunk at the isthmus, or whether you make an anastomosis there. Do you rely on the downstream seal?DR HAVERICH: The movie was inadvertently cut off. What we do is an anastomosis in the proximal portion of the descending thoracic aorta, so it is sealed at that point. That is very important, I think, because you are never sure whether you actually get complete sealing of that graft in the distal portion.DR GRIEPP: Thank you. Let me just add that of course another alternative we didn’t discuss in this session is that after putting in the elephant trunk, one can complete the repair during the same hospitalization with a stent introduced retrograde through the femoral artery. It is a little more troublesome, but it is another approach. I think you need to get Axel’s clamp, because the graft is a long thing, and that clamp looked like the way to get the elephant trunk into the arch.DR GEORGE TOLIS, JR (New York, NY): I have a question also for Dr Haverich regarding the frozen elephant trunk technique. You showed a picture where you deployed the stent in a chronically dissected descending aorta. First, do you size the stent based on the true lumen or the false lumen? And if you size it based on the true lumen, do you ever have trouble finding a stent that is small enough to fit the 10% oversize criteria that the endograft companies recommend? And do you feel comfortable deploying a stent in the true lumen in a chronically dissected aorta without worrying about compromise of potentially false lumen-dependent structures in the abdominal viscera? Have you had any trouble doing that?DR HAVERICH: Well, we haven’t had any problems with downstream thrombosis, at least in 4 patients with acute dissection. The situation is not different from what we have seen in cases where we did a conventional elephant trunk, which was usually the case if we had a tear very far distal in the arch that made it necessary to also replace that portion of the aorta.The sizing is at this point pretty much of prototypes. It is a small company that produces the stents, and at this point, for the setting of acute dissection in emergencies, the stent comes in three sizes: small, medium, and large. If the patient with acute dissection is a normal size adult male, we use the smallest one, and that is usually just a little bit bigger than the true lumen. And of course we do insert the graft into the true lumen and don’t care about the false lumen at that point.DR TOLIS: I may have misspoken: I meant to ask about chronic dissection, not acute. In acute dissection I guess it would be less of a problem, because up to a few hours before the intervention, the whole abdominal system was getting blood flow through the true lumen. In chronic dissections, are you concerned about taking out the false lumen distally and causing malperfusion?DR HAVERICH: No, we are not, because, especially in that setting, we do have experience with the conventional elephant trunk, and that also very often led to full expansion of the elephant trunk in the true lumen. So there is no difference between the conventional and the stent grafted elephant trunk in that setting.DR TOLIS: Is that what the panel thinks, too, about putting the elephant trunk in the true lumen? Or do you fenestrate and put it right above your fenestration?DR KAZUI: I want to ask a question of Dr Haverich, because there are many Japanese surgeons or hospitals where they tried to put the frozen elephant trunk in with open stent grafting at the time of total aortic arch replacement. But some of the literature said there are triple rates of paraplegia or paraparesis, and they couldn’t figure out what is causing that with regard to the size of the stent. So most of our Japanese surgeons have given up the use of the frozen stent graft. Could you make any comment on this? How can one avoid this catastrophic complication, and why don’t you have so many terrible results at your hospital?DR HAVERICH: I think most of the data that are published refer to the setting of acute aortic dissection. The length of the graft in the descending aorta is critically important. This is why we never go longer than 10 to 12 cm, and it is probably better to stick with 8 cm. In effect, if you have a small true lumen in the setting of acute dissection, you have the same effect as with a conventional elephant trunk. You get thrombosis around the true lumen, in the false lumen of the dissection. And even without using the elephant trunk, just doing a proximal aortic arch repair in the setting of acute dissection, we do see paraplegia. I think we need a larger number of patients where the techniques have been applied to further understand the pathophysiology. In our patients, there was not a single case of paraplegia. So we have to wait and see.DR RAJ K. BOSE (Tuscon, AZ): My first question is for Dr Takamoto. What is the device that you use for creating this intermittent pressure augmentation for your retrograde cerebral perfusion?DR TAKAMOTO: We don’t use any special device, just a regular cardiopulmonary bypass pump. The technician manages the knob of the flow, and increases the flow and then monitors the pressure, and raises the pressure from 15 mm Hg to 45 mm Hg. After reaching 45 mm Hg, he turns it down to 15 mm Hg every 30 seconds. We don’t use any device; the perfusionist does it.DR BOSE: The second question is to Dr Kouchoukos; this is a technical question. During the clamshell procedure, you do the arch first and I think you do the distal anastomosis second and then the proximal anastomosis last. Is that correct?DR KOUCHOUKOS: That’s correct.DR BOSE: For doing the distal all the way down in the diaphragm, which intercostal space do you go through?DR KOUCHOUKOS: For all the procedures, we use the fourth intercostal space bilaterally.DR BOSE: And if you need to go further down for access you just take out extra ribs?DR KOUCHOUKOS: We have not had to do that if the resection is confined to the thoracic aorta. It is not difficult, by increasing the depth of the incision into the axillary fossa, to get enough exposure to reach the diaphragm. So we have not used another intercostal space incision.DR BOSE: For doing these procedures in chronic dissecting aneurysms, I assume you cut the septum all the way to your distal anastomosis and you ligate all the segmental arteries. Have you had any problems with paraplegia doing this for the chronic dissecting aneurysms where these segmentals are really pretty well perfused?DR KOUCHOUKOS: No. We can sometimes bevel the distal anastomosis to try to preserve some of the distal intercostal arteries, but for the patients who have had chronic dissection where we have done the anastomosis into the distal descending or near the aortic hiatus, we have not had paraplegia.DR MICHAEL E. GORTON (Kansas City, MO): I do normothermic arch resections, and have now done 15 without neurologic injury. It is a world where we can have dissenting opinions; and you are absolutely right, we need to keep doing those to establish series large enough to compare with series that you folks have done over your entire careers.I also use a stent graft using a Seldinger technique into the descending thoracic aorta. In the United States, all we have is the TAG stent graft, and my question to Dr Haverich is how do you connect your arch graft to your descending stent graft? I saw a picture and I wasn’t exactly sure how you suture the two together. Do you actually deploy the stent within another graft and then do a graft-to-graft anastomosis?DR HAVERICH: Unfortunately, the movie was cut short at that point. We do the anastomosis in a conventional manner with the prototypes that we have used thus far. In the conventional situation, we use a running suture between the proximal descending thoracic aorta and the stent graft in its proximal segment. You can either attach your arch graft at the same suture line, or you can do it simultaneously, attaching the descending aorta, the stent graft, and the arch prosthesis in a single anastomosis.DR GORTON: With the TAG graft being fairly thin PTFE, what I have done is put in an inner and outer layer of felt and also incorporated the aorta into that anastomosis. That eliminates a type 2 endoleak if there is a patent bronchial vessel.DR HAVERICH: But if you take the aorta into the suture line, you don’t need felt.DR SPIELVOGEL: I would just like to make a comment related to the frozen elephant trunk just so people don’t forget that you can actually still do the second-stage retrograde with a stent graft into the elephant trunk from below, and this works very well. So the idea of connecting grafts to stent grafts is a good one, but just keep in mind that this is not the only way to do it. We have done a number of patients in whom we have constructed the elephant trunk and then gone in from below. This works very well, and you can accurately deploy the stent in the elephant trunk and bring it down as far as you need based on aortography or transesophageal echo.DR KOUCHOUKOS: Getting back to the issue of single-stage versus two-stage procedure, remember, at least in the United States, that sticking a Gore TAG graft into the descending thoracic aorta is going to cost you a minimum of $10,000, and possibly more. So I think one should also put that into the equation, at least in the United States, in thinking about replacing the descending thoracic aorta with a stent graft versus a conventional piece of graft, which will cost several hundred dollars.And let me just say something about normothermic perfusion. I think if someone is very careful and selective in the individuals in whom he chooses to use normothermic perfusion, he will probably have success. But if you get into a situation where the period of time, for technical problems or for other reasons, is prolonged, you run the risk, as Dr Bonser said, of a calamity. So I would urge those of you who are going to use normothermia to use it cautiously and selectively until we get more information.DR SANTI TRIMARCHI (Milano, Italy): I have a question for Professor Kazui. You mentioned that increasing age was an independent predictor of in-hospital mortality. Would you explain better this concept of increasing age? Do you have any limitation in surgical indication based just on the age of patients? Do you have any data about mortality and stroke rate in patients over 75 or over 80?DR KAZUI: In this series, aging is an independent predictor of in-hospital mortality, but these patients have serious comorbidity, including COPD, coronary artery disease, stroke and diabetes, and chronic renal failure. That is why these patients have high in-hospital mortality. As far as age is concerned, our oldest patient is 92 years old; he received a total arch replacement, and was discharged without any neurological complications. So our exclusion criteria include patients with severe stroke who are bedridden and those who have severe myocardial dysfunction. Our final exclusion criterion is chronic obstructive pulmonary disease: if forced expiratory volume in 1 second is less than 1,000 ccs, we exclude these patients from arch repair.As far as renal dysfunction is concerned, if the patient is already on hemodialysis, we can do an arch operation without any serious problem. In someone not on dialysis, if the creatinine clearance is less than 30 ccs per minute and the patient really needs the operation, we introduce hemodialysis first, and we take them to the operating room after 1 month of introductory hemodialysis.DR GRIEPP: I will just add that we reported our experience in octogenarians a few years ago, for procedures requiring interruption of cerebral circulation involving HCA. In the octogenarians, elective procedures through a sternotomy only carried a risk of adverse outcome of 3% to 5%. Emergency procedures had an adverse outcome of 30% to 35%. In contrast, if we had to do the operation through a left thoracotomy, the elective patients requiring HCA who were octogenarians had an adverse outcome of 30% to 35%, but in emergency patients, it was 80%. As a consequence of that experience, we offer emergency operations that require hypothermic circulatory arrest or selective cerebral perfusion to people older than 80 if they can be done via a sternotomy, but not if a left thoracotomy will be required. I think that probably is a group in whom the operation is just too much.DR JEAN E. BACHET (Paris, France): I would like to make a small comment about normothermia and the question raised a few minutes ago by Dr Touati. This technique, in addition to what has been said, precludes the use of an open distal anastomosis: you cannot imagine normothermia and circulatory arrest in the lower part of the body, with the liver, kidneys, et cetera not perfused for 30 or 45 minutes. So you have to perfuse distally and use either femoral cannulation, or if you are braver, cross-clamping of the aorta, and this may result in catastrophe. So I think we should be very cautious and keep on using hypothermia.My second point is about a technica