Ann Thorac Surg 1999;67:1891-1894
© 1999 The Society of Thoracic Surgeons
Discussion: session 3—aortic arch I
Randall B. Griepp, MD, Moderator,
Jean Bachet, MD, Panelist,
William A. Baumgartner, MD, Panelist,
G. Michael Deeb, MD, Panelist,
M. Arisan Ergin, MD, PhD, Panelist,
Yuichi Ueda, MD, Panelist
Presented at the Aortic Surgery Symposium VI, April 30–May 1, 1998, New York, NY.
 |
Dr Randall B. Griepp
|
|---|
(New York, NY): Dr Baumgartner, with regard to the nitric oxide synthase (NOS) inhibitor: you didn’t say anything about whether there was clinical or physiological improvement in the dogs that were so treated. Is that something that has any hope for use in human beings?
|
Dr William A. Baumgartner
|
|---|
(Baltimore, MD): The dogs were functionally improved at 72 hours with the NOS inhibitor. We have tried to use other NOS inhibitors that are more global, and they actually cause harm as opposed to help, I think because of the vasodilatation that results. Is it a drug that we can use clinically? Unfortunately, no: 7-nitroindazole is a drug that has never been tested in humans. Since cerebral ischemic injury, stroke in general, is such a high-profile problem, there are a number of pharmaceutical companies that are pursuing these types of drugs, so hopefully one will turn out to be helpful clinically.
|
Dr Griepp
|
|---|
Dr Ueda and Dr Deeb: You both use retrograde perfusion seemingly pretty much routinely, but the pressures at which you perfuse are quite substantially different, I think 15 mm Hg versus 40. Maybe each of you would like to tell us what is the basis for using the pressure that you have chosen.
|
Dr Yuichi Ueda
|
|---|
(Tenri, Japan): Usually I use 20 or 15 mm Hg. Because of our experience in congenital heart surgery, in particular with right heart bypass in the Fontan operation, we have seen that cerebral venous pressure usually is acceptable at 20 mm Hg or less. So, usually we use 20 mm Hg for retrograde perfusion. There are several animal investigations in which the maximal efficacy of brain perfusion occurs at 25 mm Hg. A much higher retrograde perfusion pressure produces brain edema. So I believe that less than 20 mm Hg is much safer than 25 or 40 mm Hg. I know there are several venous valves in the internal jugular vein, but there are very rich collaterals around the neck vessels, so I don’t require high venous pressure to produce effective cerebral retrograde perfusion: the collaterals seem to work well with a 20-mm Hg pressure.
|
Dr G. Michael Deeb
|
|---|
(Ann Arbor, MI): I don’t necessarily aim for a particular pressure. I just aim for a volume, trying to keep the pressure less than 40 mm Hg. And I don’t know that there are any really hard data in the literature that are going to show exactly what the maximum pressure is that you can use. Clearly, by increasing the venous pressure to really high levels in animal models, you can cause damage, but we have not seen any significant damage clinically with a pressure in the 30s. I don’t know what our average pressure is, but I’ll accept anything less than 40 mm Hg to get the flows that I desire. I think I judge more by flow than anything else. I don’t know about you, but in our institution most of these cases are being done in the middle of the night, and we’re putting the patient in Trendelenburg, and before you start out, when you turn the pump off, your pressure is already reading 10 mm Hg. So I’m not so sure of the accuracy of these venous pressure measurements. I try to make sure that the pressure goes no more than three times higher than what I start out with.
|
Dr Griepp
|
|---|
We always try to get the anesthesiologists to move the transducers when we tilt the table. It’s sometimes difficult.
Dr Ergin, would you like to say anything about whether patients perfused retrograde wake up more promptly than patients who are not perfused retrograde?
|
Dr M. Arisan Ergin
|
|---|
(New York, NY): In our experience they certainly don’t. And when we looked at retrograde perfusion, trying to find out whether it did any good in preventing strokes, we found that having used retrograde is a marker for stroke or a neurological injury. But I must qualify that answer by saying that we usually use it in high-risk patients, so retrograde perfusion probably is associated with stroke because of patient selection; we use it in higher risk patients. I’m not saying retrograde perfusion causes stroke, but it certainly didn’t prevent it in our patients.
|
Dr Jay I. LaBourene
|
|---|
(San Francisco, CA): Do you have any problems using aprotinin with circulatory arrest? Some people are concerned with causing hypercoagulability at cold temperatures.
Secondly, traditional teaching has been for keeping the hematocrit quite low, in the high teens to low 20s, and some recent papers by Richard Jonas from the Children’s Hospital in Boston, who has done some very elegant animal work, have shown both histological and biological results that are better with hematocrits of approximately 30.
|
Dr Deeb
|
|---|
I’ll comment a little about the aprotinin. We had a paper in the Journal of Cardiac Anesthesiology in which we compared patients undergoing hypothermic circulatory arrest who had aprotinin and amicar versus those who had none. We looked at the amount of bleeding, the amount of blood products received, and whether there was a difference in the neurologic outcome. And there really wasn’t. So we decided not to use aprotinin, not for medical reasons, but because aprotinin is quite expensive and we were unable to show any benefit from using it.
|
Dr Baumgartner
|
|---|
We have used aprotinin with patients undergoing hypothermic circulatory arrest, but not routinely. We use Amicar. But in reoperations we have used aprotinin in a handful of patients. We haven’t seen a problem, but maybe we just haven’t used it enough.
|
Dr Jean Bachet
|
|---|
(Suresnes, France): We also use aprotinin in the patient at risk, that is, in redo procedures and total arch replacements, and we do not share the experiences that have been published so many times concerning the problems associated with aprotinin in those patients. Our experience may be different from that involving patients undergoing a huge thoracoabdominal replacement in deep hypothermia, because the risk of bleeding in those patients is much more important.
|
Dr Griepp
|
|---|
Does anyone want to comment on hematocrit during cooling, during warming? If you look at those papers by Dr Jonas, they are very interesting, and I think there are lessons to be learned from them. But I suspect that the major problem with low hematocrits doesn’t have to do with the low temperatures themselves: it is on the way up and the way down that the brain is not being adequately supplied with oxygen, particularly during warming. I think there is a particular concern among our colleagues that do congenital heart surgery, because, as you probably know, the technique du jour right now is modified ultrafiltration, in which you leave the hematocrit low during the time you warm and then suck all the fluid out after you have come off bypass and get the hematocrit back up at that time. That is probably a dangerous way of perfusing children, particularly when you are perfusing with blood temperatures in the normothermic range, and I suspect that insisting on an adequate hematocrit is an important point in that setting. It is hard to believe that during reasonably profound hypothermia, below 20 degrees or so, that hematocrits above 30 are necessary. If you really want to know the answer, what you should be doing is measuring the jugular venous saturation, because that’s the only way you really know whether the brain is being adequately supplied with oxygen. If you measure jugular venous saturations, either with a little oximeter catheter, which is not quite so precise, or just by doing periodic samples, you will find situations in which the jugular venous saturation is much less than you thought it should be. But if you don’t measure things, you can’t recognize problems or figure out how to treat them. So if you monitor jugular venous saturation, then you will know when your hematocrit is inadequate, when your pressure is too low, et cetera. By measuring jugular venous saturations both during cooling and, perhaps even more importantly, during warming, imbalances between oxygen supply and demand in the brain can be minimized.
|
Dr Wim J. Morshuis
|
|---|
(Nieuwegein, the Netherlands): We, like Dr Bachet, also favor selective antegrade cerebral perfusion, although our technique is somewhat different. I have understood that during circulatory arrest, a core temperature of 28°C is maintained. Considering that the neuronal tissue of the spinal cord is as vulnerable to ischemia as brain tissue, is it not possible that the three cases of paraplegia that Dr Bachet mentioned are due to a period of relative warm ischemia of the spinal cord?
|
Dr Bachet
|
|---|
Core cooling is stopped at 28°C. But as we perfuse the brain with blood at a very much lower temperature, core temperature goes down a little, to 20, 25, or 24°C. To answer your question precisely, we are absolutely certain that the paraplegia was not linked to inadequate cooling. In two cases, I closed a critical reentry site in a type 3 or type B chronic dissection in which the spinal cord was obviously vascularized through the false channel, and since there was no reentry, the two patients became paraplegic. The third case of paraplegia was linked to ligation of the subclavian artery, which was giving off collateral arteries that were part of spinal cord vascularization. Our patients have not experienced trouble in the spinal cord or in renal function with core circulatory arrest at this temperature. Circulatory arrest duration is limited to the distal anastomosis, so the mean time of circulatory arrest is less than half an hour in this study. This is not a time sufficient to have big problems with the spinal cord.
|
Dr Griepp
|
|---|
Dr Bachet, you don’t measure the pressure in the left carotid circulation, or perhaps you do on the right side if you have a right radial artery catheter. Why don’t you? Do you think it’s safe to do so? Have you ever monitored jugular venous bulb saturations in these patients so you know what the venous saturation is during selective perfusion?
|
Dr Bachet
|
|---|
Concerning the pressure, we did monitor it routinely at the beginning of our experience, and the perfusionists noticed that the pressure measured in the carotid arteries corresponded to a certain pressure at the exit of the roller pump of the heat exchanger. So now we measure the pressure at the roller pump and we infer that it corresponds to proper carotid perfusion. We always have a radial artery measurement, and on the right side we know that the radial artery pressure corresponds more or less to right carotid artery perfusion pressure, because on the right side we almost always cannulate the innominate artery and not the carotid artery itself. Concerning the venous saturation, we don’t measure it.
|
Dr Robert A. Dion
|
|---|
(Brussels, Belgium): Dr Ergin, I was a little bit surprised that you would limit the duration of circulatory arrest to 25 minutes. In our experience the length of circulatory arrest has no correlation with any temporary neurologic dysfunction. And we don’t use any adjuncts in circulatory arrest. We base the level of temperature for circulatory arrest on the disappearance of somatosensory-evoked potentials. How did you determine the level of your hypothermia? Don’t you think that the reason why you have such a correlation between the duration of circulatory arrest and temporary dysfunction may be related to a higher level of temperature for circulatory arrest? How do you decide that in some patients you will stop cooling at 12°C and in some patients you stop at 18°C?
|
Dr Ergin
|
|---|
These data are from a study that has been going on for a long time. During the whole period, we had intervals during which we measured EEGs and somatosensory potentials, but in some patients we measured nothing. But from the beginning, we have always advocated that colder is better. I don’t think anybody cools as much as we do, and we pay a lot of attention to keeping the brain cold during the period of circulatory arrest. I didn’t mean to imply that we should limit the circulatory arrest time to 25 minutes. In this study, a cutoff point of 25 minutes was chosen arbitrarily, but it happens to coincide with estimates of safety from metabolic data. I think it’s important to try to limit circulatory arrest times, especially in older patients. Probably 30 or 35 minutes should be the limit. If you exceed that, you are going to pay a price for it. It may be only in terms of temporary dysfunction. I don’t think stroke is related to circulatory arrest time. We have never found a correlation between circulatory arrest time and stroke.
|
Dr Griepp
|
|---|
If I may add to that, Dr Dion, most or almost all of these patients had jugular venous saturations measured, and they were greater than 95% at the time the circulation was interrupted. In addition, all patients who had prolonged circulatory arrest times, or circulatory arrest times anticipated to last more than about 20 minutes, had their heads packed in ice. So the brains of these patients were very cold.
|
Dr Hideo Adachi
|
|---|
(Omiya, Japan): We usually use antegrade cerebral perfusion with cold blood, like Dr Bachet. In 100 experiences in elective surgical cases, the mortality rate was less than 10% and very few patients presented with stroke. Dr Bachet mentioned a mortality of 14.7%. I would like to know the difference between the elective surgical cases and the emergent surgical situations, especially in mortality and morbidity.
|
Dr Bachet
|
|---|
Obviously the mortality rate in emergency cases is much higher. As you say, in elective cases, it’s less than 10%. But I don’t think that the mode of cerebral protection is involved in this mortality rate. Emergency patients are mainly acute dissection patients, and we know that those patients die of anything, including strokes: they have multi-organ failure, they have malperfusion, they have infections. So although the mortality rate is much higher in emergency cases, I don’t think that it is linked with the type of cerebral protection.
|
Dr Adachi
|
|---|
I think our average cerebral perfusion time is 90 minutes, and I think antegrade perfusion is a very reliable method.
|
Dr Allen S. Hudspeth
|
|---|
(Winston-Salem, NC): Dr Deeb, if I have the right atrium open and put in 500 cc of antegrade cardioplegia, I see 400 or 500 cc of blood come out of the coronary sinus. Could you tell us how much blood you have measured coming out of these brachiocephalic vessels in retrograde cerebral perfusion? People have demonstrated to me and I have seen movies, and they say, "See, there’s black blood coming out of there, so therefore it’s going through the brain." It looks as though two mosquitoes could drink all of it that I see coming out. Do you know where this blood goes? And if you put 1,000 cc into the retrograde circuit, how much do you measure coming out of the brachiocephalic vessels?
|
Dr Deeb
|
|---|
This much I can tell you: if you don’t have a sucker down there, you’re not going to be able to see to do your surgery: the whole field is going to fill up with dark blood. If I had a patient where I didn’t see blood coming out of the head vessels, I would be severely worried that I wasn’t delivering retrograde cerebral perfusion. I have not had a patient in whom I haven’t had to have a sucker down there. I use a flexible pump sucker that maintains the position you put it in: the really flexible ones always get knocked away and then in two seconds you can’t see. If you are not seeing that blood coming back, then you are not delivering effective retrograde perfusion. I don’t think all of that blood goes to the brain: it goes into anything that drains from the SVC, so some of it is probably going to the upper extremities. A lot of that blood goes down the azygos. During operations in cases of renal tumors which extend up into the heart, when I first went to take large thrombi out, all of a sudden the entire inferior vena cava would fill with blood, dark blood, and the only blood that was going into this patient was retrograde cerebral perfusion. It took me a little while to figure out that the retrograde perfusion was going down the azygos. So now when I’m doing these renal cases I cinch the azygos. You can ask "Why don’t you cinch the azygos for all of your cases in whom you want cerebral perfusion?" The bottom line is that I don’t want just retrograde cerebral perfusion. I believe that I can also get some blood to the spinal cord, which we are all concerned about, and the liver by perfusing retrograde. If you arrest for 90 minutes, you may have been cold when you started out, but your liver, isn’t cold after 90 minutes if you’re not perfusing it in some way. So you’re going to have metabolic problems which translate into a bad coagulopathy. So I don’t think it’s all that bad to have some of that blood going down the azygos vein, getting to the visceral organs and the spinal cord. But if you don’t see blood coming back from your head vessels in enough volume that you need a major pump sucker there, then I think you’re not perfusing the brain and you have a problem.
|
Dr Griepp
|
|---|
I’ll tell you my opinion, although it wasn’t solicited. Measuring of the return through the cerebral vessels in man during retrograde perfusion has almost never been reported. The little data that do exist show return of about 10% to a high of 20% of what is infused. In the experimental animal, at least in the pig, about 80% ends up in the inferior vena cava, and 20% probably does perfuse some sort of capillary bed in the head. But I can’t believe that Dr Deeb can’t tell the difference between 20 cc per minute and a liter per minute.
|
Dr Deeb
|
|---|
I don’t think I’m getting a liter of blood in my field, but Dr Hudspeth is saying that he measures it in trickles, and I’m saying that you see more than that. The big question is: what does retrograde perfusion do? Does it really get through the venous system into the capillaries, deliver oxygen, and come back the arteries, or is it circulating around in the cerebral sinus system and in the venous system? That’s a debatable subject, and you can put up five or six articles which say it really doesn’t deliver an oxygen substrate and it really doesn’t protect the brain, and you’ll get other papers that say it does perfuse the brain. There are a lot of studies that have not been published, unfortunately, that show that retrograde perfusion actually doesn’t get across the sinusoids, that the blood doesn’t get to the arterial side. But it does fill the venous sinuses, and I think that’s important because I think that you are keeping the brain cold. Whether retrograde perfusion does deliver oxygen or metabolic substrate I’m not certain, but I do think that it keeps the brain cold and prevents embolic phenomena including particulate matter, and therefore I think it’s important. I do think that you can tell the difference between a liter of blood coming back and 200 cc, but I don’t measure it. I’m not that concerned whether the whole liter of retrograde perfusion comes back, and I don’t believe the whole liter goes to the brain. I know that enough does to keep the brain cold to give me the time that I need to do the procedure. I would be concerned if I weren’t seeing any blood coming back. I may then not be protecting the patients by preventing air and particulate emboli when I reinstitute antegrade perfusion, and I wouldn’t be keeping them cold.
|
Dr Griepp
|
|---|
There is a lot of experimental literature on this subject and not all of it is so difficult to understand. It has been demonstrated repeatedly that there is some nutritive function delivered with retrograde cerebral perfusion, and we have been able to show in the pig that there is a measurable and a significant amount of oxygen delivery to the brain, but it’s about 10% to 20% of what is needed at 20 degrees. If you take the temperature down to 10 degrees, a higher proportion of the basal metabolic needs of the brain are met by retrograde perfusion at that temperature, but it’s still not the full amount.
|
Dr Deeb
|
|---|
How are you perfusing the brain retrograde?
|
Dr Griepp
|
|---|
We perfuse at a sagittal sinus pressure of about 20 to 25 mm Hg in the experimental animals, at a temperature of 20°C. We perfuse humans at 10°C.
|
Dr Deeb
|
|---|
So even though the amount of the substrate or oxygen that you are delivering is not that great, at 10 degrees it is a significant level.
|
Dr Griepp
|
|---|
If it’s about 10% at 20 degrees, it will be about 20% at 10 degrees. These are percentages of basal metabolic requirements: there is good experimental data on this question. I think what we are debating, or what we as surgeons are trying to understand, is whether retrograde perfusion is a useful technique. My own feeling is that it is useful, but it is not a guarantee that you can turn off the pump and go on operating forever without restoring antegrade perfusion. Although I’m sure nobody on the panel would feel that way, occasionally when retrograde perfusion gets debated, such excessive optimism comes across as one point of view.
I would point out one other interesting thing. If you really want to be sure that all of the blood that you put in retrograde goes through a capillary bed, you can occlude the inferior vena cava, and we have done that in experimental animals and in man. If you occlude the inferior vena cava and perfuse retrograde into the superior vena cava, you won’t be able to flow at more than about 200 or at most 500 cc per minute without pushing the venous pressure way up, because the blood has no place to go but to traverse a capillary bed. You will sequester a fair amount of fluid in the patient, however, because you will extravasate fluid out through the capillaries. But you can prevent the runoff into the inferior vena cava simply by snaring it. I had hoped at one time that retrograde perfusion with the IVC occluded would be clinically superior. I’m not sure that it is, and experimentally we’re still looking.
|
Dr Howard H. Shiang
|
|---|
(New York, NY): I would like to ask, Dr Ueda, how much flow you measure from retrograde perfusion to the brain.
|
Dr Ueda
|
|---|
I don’t know exactly what volume is going into the brain, but my technique suggests that pressure is a more important factor limiting cerebral perfusion. The flow depends on the pressure. As far as I know from my experience, the flow rate varies from 400 to 1000 cc/min. I know that it has been published that the percentage of flow that goes into the brain should be a third or a quarter of retrograde perfusion.
This article has been cited by other articles:
|
|
|
|
 
M. A. Pichlmaier, O. E. Teebken, H. Baraki, and A. Haverich
The frozen elephant trunk technique
MMCTS,
January 1, 2007;
2007(0329):
mmcts.2006.001990 - mmcts.2006.001990.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
F. F. Immer, C. Lippeck, H. Barmettler, P. A. Berdat, F. S. Eckstein, B. Kipfer, H. Saner, J. Schmidli, and T. P. Carrel
Improvement of Quality of Life After Surgery on the Thoracic Aorta: Effect of Antegrade Cerebral Perfusion and Short Duration of Deep Hypothermic Circulatory Arrest
Circulation,
September 14, 2004;
110(11_suppl_1):
II-250 - II-255.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
D. Spielvogel, M. N. Mathur, and R. B. Griepp
Aneurysms of the Aortic Arch
Card. Surg. Adult,
January 1, 2003;
2(2003):
1149 - 1168.
[Full Text]
|
|
|
Top
Dr Randall B. Griepp
Dr William A. Baumgartner