| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2001;71:S144-S146
© 2001 The Society of Thoracic Surgeons
Presented at the Fifth International Conference on Circulatory Support Devices for Severe Cardiac Failure, New York, NY, Sept 15–17, 2000.
DR ROBERT L. KORMOS (Pittsburgh, PA):
I would like to open up discussion now on these devices. I am sure we all have a number of significant questions. We have about 15 or 20 minutes, so I am going to open it up to the floor.
While people formulate questions, what I would like to ask the panel members to tell us your thoughts about response to exercise with these devices. Because clearly we have devices that are going to be adjusted by the patient individually, we have others that presume that you do not have to adjust them, and some that will have a sophisticated control mechanism. So I guess what I am asking is, how important is a control mechanism ultimately? And second of all, if there is no control device, what are the implications of the patient adjusting his or her own device? Maybe we can start with Ernst.
DR ERNST WOLNER (Vienna, Austria):
In my opinion this type of device could be an alternative to transplantation in the future, which is the ultimate goal of this investigation. I believe the pumps have limited styling mechanism when the preload is high or the pump or heart is functioning better, which results in uneven flow. I would think some sort of physiologic control would be necessary, for example as in some pacemakers. We have pacemakers in which the frequency increases with movement and conversely decreases when there is limited movement as in sleeping. So perhaps this could be one solution or, as discussed by Bud Frazier, controlling it from the pulsatility.
From our experience, especially in the initial period in sick patients, there was no pulsatility. Pulsatility comes back after a few weeks, when there is some recovery of the heart. In my opinion, when these devices are used as an alternative to transplantation, then some control mechanisms are necessary.
DR GEORGE P. NOON (Houston, TX):
What we have seen in these patients is that the exercise is determined by first the right ventricular function, and second the left ventricular function, with whatever comes out the aortic valve and whatever your pump is pumping. And there is no question that in a patient, if you do increase the rpm through the pump, then you can increase the total output of both. But what we have seen is that there are patients who, without changing the pump, can increase their exercise up to 100 watts with no changes in the pump. And then we have seen other patients whose left and right ventricles are not working well, and when they exercise, they have a problem. Increasing the rpm in those patients is not going to make a significant differ-ence in their ability to exercise because they are not getting blood over from the right side.
So the patient who has a fairly good functioning heart can get by very well without changing the rpm once it is set at a good range.
DR KORMOS:
Doctor Frazier.
DR O. H. FRAZIER (Houston, TX):
When we anesthetize patients with advanced heart failure for device implantation, they are alive, but with limited cardiac output. If the cardiac output were not compromised, we obviously would not do the implant. Our average preoperative cardiac index has been about 1.4, but all the patients in the current protocol for bridge-to-transplant with the Jarvik 2000 have left the operating room without pulsatile flow. Based on this experience, you do not need higher flows. If the flow is too high, the ventricle can literally be sucked down, which can damage the heart muscle and cause hemolysis or thrombosis. You may have this complication if you totally unload the ventricle.
All these patients, barring complications, will improve with time, whether they have a TCI pump or any other assist device. Their exercise tolerance will also increase.
Physiologic control mechanisms can certainly be used. We did this in our early NIH-funded series, and the other contractors probably did it also. We controlled the pump with the heart rate. You can also control it with pressure. There are a lot of other ways.
What we should be concentrating on now is saving these patients and getting them out of the hospital.
DR KORMOS:
Doctor Griffith.
DR BARTLEY P. GRIFFITH (Pittsburgh, PA):
Our experience, primarily with the hemopump, or the HeartMate II, has been with calves. And I can tell you, calves are not smart enough to turn up the pump when they are on the treadmill, so you really need an attendant. I think quality of life and all those issues have been discussed to the extent that, clearly, as patients wean from their doctors, they probably ought to have an auto-control mechanism that fits their physiologic needs. I would not argue with Dr Frazier’s point that less is best. And that can be decided clearly by the pulsatility index of the left ventricle, which is basically current draw, if you wish, which correlates directly with flow, at least in the HeartMate II device. So there are ways, without placing sensors in the heart, to get some idea in terms of the filling. And left heart filling will be dependent on right heart delivery, George. If the right heart is healthy and the patient is active, then you want to pump more, and I am sure everybody would agree with that.
I do agree with Bud that first things first, and I think it is great to crawl before you walk, and before you walk maybe stumble along before you run. But ultimately our patients are going to want to run and I think we are going to need some type of control of these pumps that helps them do that.
DR KORMOS:
Jim, how do you control your pump? Because you have virtually the same problem. I mean, you have the same potential of sucking the heart down and having a sort of runaway pump syndrome. How do you control it?
DR JAMES A. MAGOVERN (Pittsburgh, PA):
Basically, we have gone by clinical measurements. In our setting the heart is pretty sick and so initially there is no pulsatility. Some pulsatility is preferable. As the heart starts to recover you do get pulsatility. The patients have a Swan-Ganz catheter in place and you try to keep the pulmonary artery and the diastolic pressure at a reasonable level. But basically control has just been on clinical grounds.
Another way to do this is by changing the pump rpm and then watching flow. Initially, as the rpm’s go up, the flow and the blood pressure go up. At a certain point, the rpm’s go up and the blood pressure does not, and then if you continue to raise the rpm’s the blood pressure goes down again. That is "the poor man’s way" of doing it. But realistically our pumps flow around 3.5 to 4 liters in most of the patients, and we are hoping to get another liter or two out of the patient.
DR KORMOS:
I would like to open the discussion up to the audience. Please identify yourself and your institution.
DR GEORG M. WIESELTHALER (Vienna, Austria):
I would like to make two comments. The first concerns the necessary pump flow. We have some data ready from our patients that show you have to have a certain amount of blood flow through the pump. For the first 2 patients, as it has been shown from Dr Wolner, we had some flow restrictions. These patients had low mixed oxygen saturation, which means they had not had enough tissue perfusion. As a matter of fact, these patients could not tolerate the exercise testing we conducted in all the other patients. So I think to have good tissue perfusion, you have to have enough flow available. And if the heart is not producing that amount of flow, you have to make it happen through the pump.
The second comment concerns the need for physiologic control. In patients in whom we did spirometry and some invasive hemodynamic measurements, we have seen them run on a mixed venous oxygen saturation of 60% to 70%. When they start exercising, the mixed venous oxygen saturation goes down and even lactate increases if you keep them on a constant speed. If you increase their speed and increase the pump flow, the lactate they produce during the exercise is way lower than it is without that pump speed increase. So probably if you have a lower pump output and a lower overall cardiac output, there might be some detrimental effects during prolonged exercise in these patients.
DR ANDREW H. GOLDSTEIN (Columbus, OH):
Jim Magovern had mentioned the efficacy of atrial cannulation with the AB180. One advantage to that, of course, is the ability to place the device not only without cardiopulmonary bypass but without having to fibrillate the heart; in fact, that was our experience in the two implants that we performed at Ohio State.
The other more interesting aspect of atrial versus ventricular cannulation has to do with oxygen demand relationships of the left ventricle and the implications for long-term ventricular recovery. All of our dogma about left ventricular decompression and resting of the left ventricle are based on experiments from the big pulsatile devices or roller pumps, which, have their own unique hemodynamics. But rotary pumps are amenable to graded bypass, that is, partial bypass over a long term. I am wondering if any of the panelists have given thought to long-term use of these pumps with partial bypass and what that might mean in terms of left ventricular energetics?
DR NOON:
What do you mean by partial bypass?
DR GOLDSTEIN:
With left atrial cannulation, you can intentionally allow some of the flow to go through the pump and some of the flow to be ejected by the left ventricle. It is not as easy to do with the axial flow pumps as with centrifugal pumps. But if you were to turn your pump speed down, you would be allowing the left ventricle to do some work and allowing the pump to do some work. It sounds counterintuitive, but over the long term it may be advantageous for oxygen demand.
DR NOON:
With the axial flow pumps, first of all, atrial cannulation would not be a good idea because the delta P between the atrium and the aorta is so high that you would not have trouble getting good flows.
The other thing is that, in contrast to what Dr Magovern said, in following patients with the Bio-Medicus pump where we have carried out atrial cannulation and supported them for postcardiotomy and done imaging studies, I have not seen decompression of the ventricle like we do with ventricular cannulation. And so from that standpoint, I, at one time, thought that we should avoid ventricular cannulation and go by atrial, but I do not think that way anymore. I think ventricular cannulation is the way to go.
The other thing is that with these pumps you do have the option of adjusting the flow so that you can have the heart work on its own and have output through the pump.
DR RICHARD A. OTT (Orange, CA):
We have heard for many years the importance of early return of renal function for the ultimate success and recovery of the patients on these devices. Are there are any special considerations and limitations with these non-pulsatile devices with regard to implanting patients with significant impairment of renal function and their subsequent return and survival? Has there been any experience with patients on the higher end of renal impairment at the time of implant?
DR NOON:
I can say that renal function has been either maintained or improved in most of these patients. But similar to the pulsatile pumps, if you have patients who have significant renal dysfunction, they continue to have problems. But I do not think that we have seen any problem with renal perfusion and renal function related to these pumps.
DR OTT:
Have you done anything special to modulate mean arterial pressure with respect to return of renal function or to try and improve what you see immediately after implant?
DR NOON:
The mean arterial pressure is usually in the 80 to 90 range and that is where we keep it. The pump index is 2.5 or so, and with that we have been able to maintain good renal function.
DR WOLNER:
I would like to add a comment. A few of our patients with low cardiac output syndrome before transplantation required some renal replacement therapy and therefore we used hemofiltration. In all other cases, with the exception of a patient with myocardial infarction who died of multiorgan failure, renal function recovered. And perhaps you have overlooked one of my slides, where I showed that creatinine and blood urea nitrogen levels returned to normal values after 3 months.
We also did some studies involving these different hormones, and there was no difference.
One of the interesting findings was that, although we had assumed these patients would develop hypertension after a few months and that the blood pressure regulation would decrease, there has been no hypertension.
DR FRAZIER:
In our 4 patients, we have maintained pulsatility for 95% of their hospital course. Their renal function normalized, although it was impaired before device implantation. One of these patients was a redo coronary patient who needed several units of blood, but the other patients needed no transfusion. The short pump time (20 to 40 minutes of partial cardiopulmonary bypass), minimal incision, and minimal blood loss helped improve renal function during the postoperative period.
DR ROBERT J. CUSIMANO (Toronto, Canada):
This morning we heard that with increasing time, when you cut out the cores and when you explant the devices, there is more fibrosis in the hearts. We also heard that as time goes on there is less chance of removing the devices to recovery.
Today Bud Frazier showed us spontaneous echo contrast in the ascending aorta with no pulsatility. You have to wonder whether this spontaneous echo contrast is related to the severity of illness in these patients or whether it is related to lack of synchronization of the pulsatile devices with coronary blood flow. The heart is the only organ that requires blood flow during diastole. Do you think that coronary blood flow and perhaps recovery to function will be better with these continuous flow devices, which will allow flow during diastole, or perhaps synchronization of the pulsatile devices?
DR FRAZIER:
I do not agree with your first two statements. I think the idea that recovery is tied to time is also false. There is always some degree of recovery, and we need to know more about its mechanism. All we can do now is describe our observation—which is that some of these patients improve enough to remove the left ventricular assist devices. Every scientific study that has examined the myocardium at the time of device implantation or examined it at the time of explantation has shown that the sarcoplasmic reticulum’s handling of calcium transport, deranged in heart failure, is improved. Apoptosis also decreases. Expression of tumor necrosis factor decreases. So all of these measurements improve.
If you totally unload the ventricle, the endocardial circulation decreases. We learned this in animal studies with the Hemopump 10 years ago.
However, coronary blood flow has not been well studied, clinically. The only good study in patients is the one conducted at Columbia Presbyterian. The positron emission tomography studies conducted by this group have shown that, if the TCI pump is in the automatic mode, the aortic valve never opens, and circulation to the endocardium decreased measurably.
So there are all sorts of implications. We need subjects, more than 5 or 10. And we need combined studies with prospective implications. But nobody knows anything for sure except we have seen the wound in these patients. One patient, who did not have an alcoholic cardiomyopathy, had a Berlin device. He’s been out nearly 5 years and he is doing well.
DR GRIFFITH:
Bud, do you think pulseless perfusion will be a better way to see bridge to recovery than nongated pulsatile?
DR FRAZIER:
Pulseless perfusion is the only way to gently unload the heart and allow for recovery of the ventricular function (by retaining normal coronary circulation). This may allow for optimal recovery of ventricular function.
DR NOON:
Let me just say something. In some patients who we studied in Berlin with Dr Hetzer and his echocardiographer, we were able to see the difference in coronary flow with and without an intraaortic balloon pump. Just as you would suspect, there was a pulsatility of flow without the intraaortic balloon because of the diastolic relaxation. All these patients had pulsatile flow going through their heart, and this was well demonstrated by the echocardiographer.
DR CUSIMANO:
The only point I make is that with the nonpulsatile devices you are talking about now, you do get diastolic flow, whereas with the nongated pulsatile devices, there could be problems.
DR KORMOS:
I think we will have to continue this. This could go on for hours. We do have to take a break, and I am sure that no one else has the guts or other anatomy to dare ask another question.
I would like to thank our speakers for an entertaining panel and for your presence. Thank you.
This article has been cited by other articles:
|
|
 |
|
  R. Hetzer, Y. Weng, E. V. Potapov, M. Pasic, T. Drews, M. Jurmann, E. Hennig, and J. Muller First experiences with a novel magnetically suspended axial flow left ventricular assist device Eur. J. Cardiothorac. Surg., June 1, 2004; 25(6): 964 - 970. [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 |
