Should We Expand Use of Hypothermia into Non-Out of Hospital Arrest Patients?



Summary Statement: 


Should We Expand Use of Hypothermia into Non-Out of Hospital Arrest Patients?

There has been considerable success in treating out-of-hospital arrest patients, especially in situations where the arrest is witnessed, e.g. a VF arrest, particularly if the patient gets CPR and hypothermia is initiated within an hour.  It is accepted that hypothermia is currently the gold standard for treating reperfusion injury and minimizing neurologic damage.  This suggests that hypothermia might be beneficial in other situations such as cardiac arrest in the hospital, treatment for STEMI patients being taken to the cardiac catheterization laboratory, or perhaps even in patients in asystole.  From the initial data, patient selection is an extremely important factor, as negative outcomes may be driven by co-morbid conditions in the very sick patients receiving hypothermia.

What are the Gaps in the current knowledge?

 If hypothermia were widely applied in the setting of STEMI, would outcomes be better, given the time it may take away from a shorter D2B time?  Do we have sufficient data on asystole patients to determine the real potential of hypothermia in this group? Do we know the real costs of this therapeutic approach?  What outcomes should we be measuring?  If the EMS-STEMI system were even more integrated, could this approach be more widely applied?  Should centers that want to pursue this approach be given ability to exclude patients from more publically reported data to encourage broader use?

Our Summary and Recommendations: 

Focused research is needed.  Centers and systems of excellence need to be built around the integrated use of hypothermia. Data registries need to focus on a variety of clinical, economic and quality of life assessments. Technologies need to be developed to deliver hypothermia as quickly and as efficiently as possible.




Panel Discussion. Should We Expand Use of Hypothermia Into Non-Out-of-Hospital-Arrest Patients?

Speaker: Mike Mooney
Moderator: Dick Shaw
Panelists:  Rita Watson, Tim Henry

Dick Shaw: The focus of this session is whether we should expand the use of hypothermia into non-Out-of Hospital-Arrest (OHA) patients. Dr. Mike Mooney from the Minneapolis Heart Institute Foundation is our speaker and he will frame the issue that will guide our discussion.

Mike Mooney: Thanks a lot Dick. I would like to go over some basic information to help frame our discussion. Therapeutic hypothermia is the gold standard treatment for reperfusion injury. This is true across all organ systems and across all species. It is a very powerful technology for reperfusion injury. If we consider where reperfusion injury occurs clinically, it is important in both cardiac arrest and in acute MI. In the situation of cardiac arrest, it is often both an acute MI and a neurologic injury causing anoxic encephalopathy. Now if you look back at what the injury is at the time of reperfusion, I would liken it to a shotgun blast. Many processes are happening independently: calcium influxes, sarcolemma problems and free radicals released. For that reason, no single target is likely to be effective from a pharmaco-biological perspective. Fortunately, therapeutic hypothermia is pluripotent and this is important when we consider how we might use it beyond cardiac arrest. It is currently the recommended standard of care for cardiac arrest in Europe. In the American College of Emergency Physicians and ACC Resuscitation Statement, it is listed as a 1A recommendation and in the PCI guidelines that are coming out soon it will also be listed as a 1A recommendation. It may be an emerging treatment for acute myocardial infarction that is something to pay attention to.

Therapeutic hypothermia however, is markedly underutilized in the United States and that is a problem that I believe we need to discuss. The best patient for therapeutic hypothermia is out-of-hospital with less than 15 minutes of down time that is witnessed, where the initial arrhythmia is ventricular fibrillation and that there are bystanders to do CPR.  Also these people do best if they have a STEMI, which seems paradoxical but is true. It is also probably true that it is best if the arrest occurs in Minnesota, Seattle or Arizona because the systems of care are present in these places to give care in an organized manner for these patients.

The patient is best treated if cooling is started within the first hour after the arrest. Many of you may have the technology and it is in your hospital and you have an idea of who it is ideally suited for, that is, the out-of-hospital cardiac arrest patient. But you may have an intensivist, an ER physician or a neurologist who will pull you aside and say, “I have this patient – should we use the therapy for X, Y or Z patient?”  

Our topic here is for in-hospital cardiac arrest, and the problem with in-hospital cardiac arrest is that the outcomes are terrible. They are at only ½ as good as out- of-hospital cardiac arrest. There are many, many reasons for it, but most of all it is the fact that many of these patients have multi-organ failure, and the origin of the cardiac arrest may not be cardiac itself, but it may be reflective of multi-system disease. Well here is a situation that I have been presented with over the years: difficult intubation with prolonged anoxia that is not necessarily reperfusion because there is an oxygen deficit, but there is perfusion happening and there is a pulse. We have used hypothermia with a good effect for prolonged anoxia. PEA asystole anoxia is probably the number one thing to pay attention to that brings poor outcome. Drug overdose is a hypoventilation issue. We will get asked that question regarding cooling in these patients.

There are certainly no data on that but we have had some survivals. Codes from the EP lab, incessant ventricular tachycardia and tamponade that has been hemodynamically unstable for a period of time, are often referred to us. Sepsis is uniformly a poor indication, but you will get asked that question for a patient who goes into a cardiac arrest from sepsis and be asked whether this is a reasonable consideration. So you can expand to in-hospital cardiac arrest. Certainly the best one is a STEMI that is in hospital and is otherwise uncomplicated. But as you expand outside the world of out-of-hospital cardiac arrest for VF, there is a price to pay. You have to be careful because both your successes and failures will be remembered. A more pressing issue for us to also consider, especially in light of what happened to Professor Eberhard Zeitler, who died of a cardiac arrest and was resuscitated but did not regain consciousness, is how do we advance therapeutic hypothermia as a standard of care in the United States for out-of-hospital cardiac arrest when it is so underutilized? That is the most pressing public health issue that this topic generally brings up. Can we integrate it effectively into a STEMI network? and can we, as we do that, eliminate the death penalty as we get assessed on our competencies and outcomes within our STEMI programs? Can we foster a national comprehensive CV emergency system that incorporates cardiac arrest, STEMIs, aortic dissections and AAA ruptures in a comprehensive manner so that we are effectively addressing these major catastrophes? These are things we need to think about.

Dick Shaw: Thanks Mike. Our panelists are Rita Watson and Tim Henry. Let’s begin with Rita. Rita, what do you think about the possibilities of expanding the use of hypothermia and could you comment on some of the points that Mike brought up based on your experience in your area of the country and at your center?

Rita Watson: I am from New Jersey. I work in two different institutions, each of which uses two different cooling systems. Studies have shown that cooling, even with ice bags and cold IV fluid, works. Even if it is delayed for hours it works. But I think there is a difference in how quickly the devices can be put on. In one of my institutions, we use cooling pads. These went on in seconds in someone with peri-MI incessant VT who came in absolutely cyanotic. We took him to the cath lab, and he got the cooling before the reperfusion, which is critical. He was discharged after full recovery of function in five days. At another of my institutions, it is not so easy to get all the cooling equipment together. But I think that the data show that as long as they are cooled, no matter what cooling device you use, there is a possibility of success. I think if we make it easier to get patients cooled, hypothermia will be used more often.

Tim Henry: Mike summarized it well.  I completely agree we really need to expand this concept of acute cardiovascular emergencies and it really should be a system throughout the whole United States. There is really no excuse for it not being done. It makes a huge difference when you have it in place. You can’t decide to do it at two o’clock in the morning without preparation. This has to be worked out ahead of time and these patients are incredibly complex. I can’t take care of them by myself for sure. 

To do this well you need an intensivist, a neurologist and an anesthetist. You need all the pieces in place. So I think having centers that actually take care of the full spectrum of acute cardiovascular emergencies is critically important and it makes a difference. How you expand the use of it is a really good question and we have expanded a lot beyond the initial trials. Clearly it works with cardiogenic shock. With patients transferred in, you can make a difference.

I think Bill O’Neill could talk about the issue of cardiogenic shock. I think it is difficult to show, but we believe that in our cardiogenic shock patients that the hypothermia improves the myocardial function as well. Whether you should expand that into patients who don’t have cardiac arrest is challenging. The second expansion area that I would bring up that is becoming more promising is in trauma. We exclude trauma right now but there is some evidence that maybe you should use hypothermia, but again it becomes more complex because the patients are more complex. I would be interested in getting thoughts about other areas and in particular, the cardiogenic shock issue by itself without cardiac arrest.

Mike Mooney:  I reviewed a paper for Circulation on that topic from a Swedish group and they showed hemodynamic data on patients in compensated cardiogenic shock exposed to hypothermia and indeed, their cardiac indexes did go up and they improved. They needed less inotropic support. When we started showing our data on cardiogenic shock, Judith Hochman called and asked if we could look at our data more carefully, to answer the question as to whether it as promising as an independent approach to patients in cardiogenic shock without arrest.

Bill O’Neil: Mike, that was a very nice summary and I guess the purpose of this panel was to try to figure out how to expand cooling beyond out-of-hospital cardiac arrest. I did a lot of the early trials with the PAMI group where cooling was used and in ICE IT. The results were really intriguing and we were almost two-thirds of the way there, but unfortunately the device companies ran out of money. What we found was that it took about 40 minutes to get patients optimally cooled. You have to get their body temperature under 35 degrees centigrade in order to really do anything in terms of myocardial salvage. At the same time that we were doing these trials there was this mania for door-to-balloon time.

I think it would be interesting to hear your comments, but we see a lot of people skipping steps like visualizing the non-culprit lesion because they want to save five minutes on the door-to-balloon time. Everybody is getting graded on this and your core measures are looked at, so with that mania, we couldn’t get people to put the cooling device in and wait for the 40 minutes. It really lost a lot of steam because of this. But when we did the subgroup analysis, when core body temperature of the patient was less than 35 degrees centigrade, there was a dramatic reduction in infarct size. It was a subgroup analysis, however, and things stopped when the overall study didn’t meet the primary end point and both companies ran out of money. The one other thing that I am hoping people would consider doing is that you can dramatically cool by infusing saline. That is one of the things that you can do that we didn’t do in the trials, but if you really wanted to rapidly start cooling you can give ½ a liter of iced saline intravenously. You can quickly start the temperature dropping and you can get the patients cooled within 20 minutes of the onset of putting the devices in. I am really enthused that you guys are still interested in this and I think it is still the kind of a topic that is two-thirds of the way answered. I am hoping there will be others that will continue to do these trials. I am absolutely convinced that you can significantly reduce infarct size if you can cool the core under 35 degrees before you re-perfuse the myocardium.

Dick Shaw: Bill, those were great points.  

Howard Cohen: I’d like to start by making a couple of points. First, I want to commend and acknowledge the great work that Mike Mooney has done to champion this process. I know this is a very difficult group of patients to take care of, but you have really done an amazing job being the champion of this and proving that it really is very effective. So kudos to you, Mike. One of the points that you made that I think is extremely important is that the selection of the patient is critical, because there is no question using hypothermia is resource and time intensive. If you pick the wrong patients, your success rate will go down and the cost of the failures may make it difficult to maintain a cost effective program. You have to pick the right patients or at least avoid the patients with a very low likelihood of success. It is similar to what we face in any trial, that the selection of the patients is really crucial. Another point that you made that I think is really important, and we will talk a little more about it tomorrow, is that reperfusion injury is a multisystem, systemic problem. It involves all the organs and there are clearly some paracrine type of effects that occur with elaboration of reperfusion salvage kinases. They protect a lot of organs, not just the heart and brain, but probably all tissues. Congratulations to all of you that have been working in this difficult area.

Mike Mooney: Thank you, Howard.

Kirk Garratt: I would like to hear a little bit of comment and discussion about the issue of cardiology and the cath lab in this.  We have a big program that does this, as many others do, but particularly when you have rapid activation of cath labs, either by the emergency room or by EMS systems, which happens in a number of places where you can call ahead, how should you decide which patients can go directly to the cath lab? We find situations where, regardless of the age of the patient, regardless of the downtime, regardless of the initial rhythm, regardless of anything else, it is a cardiac patient so we activate the cath lab. You can get a lot of false positive activations. On the other hand, perhaps so many of these people do have coronary disease and you possibly can help them. I would like to hear a little bit of an opinion about this issue.

Mike Mooney: Well it is a controversial topic. The current themes that have been developing in cardiology and the cardiology community interested in this area, notably Carl Kern and our colleague Mark Onock from Slovenia, is to take most patients to the cath lab, especially if their index arrhythmia was ventricular fibrillation. But there are those where the facts imply a non-ischemic origin, such as preexisting cardiomyopathy or other conditions. Even the most avid advocates really only take their patients to the cath lab 70% of the time.  So if you are taking around 70% of your cardiac arrests you are probably on the sweet spot. It said that the hypothermia itself prevents the proper interpretation of the electrocardiogram in showing ST segment elevation and you can have occult occlusions that otherwise would go unrecognized. The nice thing about being in the cath lab is that you do get a chance to stabilize in a very nice environment so that if the patient needs an intra-aortic balloon pump because they are in cardiogenic shock or Swan-Ganz monitoring, triple lumens all of that can be put into the patient in an expeditious manner. There are some patients that clearly do not need to go to the cath lab and you should not feel that you are being placed at risk by not taking them.

Steve Bailey: To extend that discussion one of the areas that I think takes a lot of work, and Mike you are to be commended on as well, is the relationship with EMS and the region. But that difficulty in deciding who comes to the cath lab often actually resides in the EMS system, clearly with refractory V-tach, and perhaps V-Fib. But how do you deal with the patient who is asystolic and what criteria are you using to decide who you wouldn’t bring forward? Question two is these are expensive patients when you get done caring for them. Not always with good return of quality of life. What discussions have you had with your facilities about the costs of those patients and the impact they have?

 Mike Mooney: I think it is your decision. There is not one formula that will work for everyone, but if you include mostly asystoles in your target population and don’t control them, you will not have good results, probably with survival rates of about 20% at the highest, and for a STEMI, survival rates are about 70%. So it does make a big difference.

Steve Bailey: So we made the decision to exclude asystoles if they don’t have return of rhythm.

Mike Mooney: You’re right.  I think you can help by how you write the protocol. It is a conscious choice to exclude, and to exclude asystole is justifiable, they were not a part of the original clinical studies or the Steven Bernard’s study from Australia. So they are not the index group, they are non VF patients and it would be justifiable to exclude them. When you do include them you have to be prepared for bad results. You need to put it before an ethics committee and get everyone used to what they are looking at and what to expect.

Steve Bailey: So with regards to the hospital facility, dollars spent on these patients, how do you deal with that?

Mike Mooney: Yes, it is really interesting as it is the question of whose dollars, Steve, because there are a lot of costs to the intensive care unit. But these patients are coming to the intensive care anyway. They have been resuscitated, the ventilator is not going to be turned off and they are going to be brought to the intensive care unit and managed for an extended period of time. Additionally, from a system point of view, it appears that the neurologic recovery with CPC grades 3 and 4 that require nursing homes, are dramatically reduced by hypothermia, and this can offset some of the cost. But there is no question it is resource intensive and it is a full on-court press and effort with multiple specialists involved to get these patients through. Fortunately, among the survivors, the quality of the survival is really outstanding, so there is like a switch effect, a categorical effect. If it works it works, it works really well, and if it doesn’t work the patients usually die.

Steve Bailey: I heard you mention one important point and that is that even though patients come in our cardiology cath lab, if they look like they are terminal that is not something that occurs in the cath lab because that has been a real concern I know in our facility and others. It sounds like they go to an intensive care unit and that is where that terminal care is provided.

Mike Mooney: Yes that is true.

Dick: We have a couple of more minutes. Larry you can go and then Jim.

Larry Dean:  Most of that benefit that we saw in Seattle over the years has been rapid access and the EMS system itself. Putting that in place and making sure these patients are rapidly resuscitated and brought to the hospital, the advantage that you see in Seattle is not because they all went to the cath lab and had some intervention done. It really was the EMS part of the care. So I can tell you we don’t routinely take patients with VF arrest to the cath lab without STEMI, so this whole issue about who do you take to the cath lab is I think part of the problem. I know there is some scattered information out there suggesting that it is beneficial. But that next step, that next expenditure that has to occur, at least in my experience, has not really paid off from the standpoint of finding something at cath we can do anything about. I think that the EMS piece is the first piece. You have to get back to what Tim was saying a minute ago and what you guys have done as well. You have to get the EMS system consistent within your community first and then this next step of taking more complex patients to the cath lab can only be addressed with better data.

Rita Watson: Does your EMS institute cooling prior to hospital?

Larry Dean: Yes and that really started with iced saline actually in the field.

Mike Mooney: I think that is a justifiable point of view. But Larry you do take STEMIs to the cath lab, right?

Larry Dean: Absolutely.

Mike Mooney: I think that is a reasonable point of view, you know Spalding has written widely that there is room to consider a more aggressive approach as Mark Onock proposes. 

Jim Zidar: In North Carolina, we had a very aggressive RACE project which really changed the focus in the EMS throughout the state, particularly in the large Charlotte area, Raleigh area, and Wake county EMS. And to go the next step, I know Chris Granger and Jamey Jolis in the Duke system have been pushing hard. It gets back to Bill’s point about reimbursement and who is going to pay for this. Blue Cross, at least in North Carolina, has been very supportive financially and I think Medtronic has also been behind it. Are you doing anything in Minnesota like we are trying the RACE project, which is exactly what you are talking about with out-of-hospital, cardiac arrest, and putting it into a clinical trial? Are you systematically in the Minneapolis area in a registry or a trial of some sort?

Mike Mooney: We don’t have a statewide trial. We are supported by Medtronic through the Minnesota resuscitation consortium and we are looking more at defining best standards of care across the state, but it is not in the form of a clinical trial. It is in the form of a cooperative working group.

Dick Shaw: But you are tracking data, right?

Mike Mooney: We are tracking data yes.

Jim Zidar: I think the hardest thing for the next step to go beyond STEMI is in the details, and specifically, the finances of this. Do we exclude all those with asystoles? How do you get to the details of this, especially if they are not cooled in the field and then you are left with this sort of conundrum in the ER. You have the whole team coming in from the cath lab, so do you just keep moving forward, do you cool them first?  There are a lot of issues that we are struggling through locally. I just wanted to hear if either Seattle, Arizona or Minnesota are doing something better. Tim maybe you are up on this.

Tim Henry:  I would like to ask a question because we are keeping track of these and Mike has done a beautiful job. We know that if you do PEA and asystole and you include them, our data says that you have about a 20% chance of survival to hospital discharge neurologically intact. Let’s take a vote: is 20% good enough? So if you knew with PEA and asystole that you could get one out of five normal when they leave the hospital, would you do it? How many would do it?

Jim: I don’t think I would do it.

Tim Henry: But the thing is that you come to that level, what is the level when you don’t do it? Is it 15%? Is it 10%? So at least you have some data, but it still becomes a difficult decision don’t you think?

Mike Mooney: Absolutely. And they may or may not have been witnessed, and they may be 85 years old. You could have an 85-year-old patient that looks even older than that who has asystole and was an un-witnessed arrest. There will be patients brought to you that you know you can’t help and so you do have to be willing to say no. The 20% survival rate is a really interesting question. We brought it before our ethics panel and they suggested we keep on helping those patients and including them in our registry, but they wanted to see more data over time. It does wear on the staff because that does mean that only 1 out of 5 will survive and it is a lot of effort and a lot of deaths in the intensive care unit. You have many, many specialists that come through seeing the patients, probably wondering why are we doing this if these patients are mostly dying.

Jim Zidar: I also think the interventional cardiologists already have a large burden of afterhours call, more than we had a decade ago, because at least in our state with the RACE project and the push, every day you are doing some STEMIs. Now if we add cardiac arrest we are pretty much going to have to get to a system in the bigger hospitals of having somebody sitting on site at night. We have that in some of the hospitals in our state where we have an interventionalist on all the time, staying in the hospital. I don’t know at what capacity, but maybe you have that in big cities like New York.

George Vetrovec: I am personally willing to do everything for any patient with an opportunity to recover, even if the odds are low such as the 1 in 5 recovery for patients with asystolic OHA because for the survivors there is a real benefit. However, the current practical issue has to do with outcome statistics. There is no special category for these patients in the NCDR and unless a program has a huge procedure volume, doing low probability of recovery patients will adversely affect a program’s statistics. The NCDR does not count death post procedure based on cause. Thus one can perform a successful PCI without complications, but if the patient has a persistent and hopeless CNS injury and care is appropriately withdrawn, the outcome data will record this as a PCI death. Reasons for this include new treatments moving faster than the NCDR’s response coupled with a “silent” national health policy attempt to put the onus of excluding high risk patients from treatment on physicians. By not appropriately accounting for risk in high risk, expensive patients health policy experts silently hope to curtail costs without policy by covertly driving doctors to not take on every patients who might benefit is possible but low probability -- in short encouraging practitioners to “Step Over Grandma at the door.”  Thus, the public needs to know that decisions for the performance of high risk procedures is not necessarily directed to giving every reasonable patient a chance, but most realize that some institutions with the best outcomes are those not participating in new, high risk procedures. Conversely, “Skimming” the most optimal patients is incentivized by outcome statistics which often effect payment and “publicly perceived” quality, which adversely affects decisions to attempt to “save” high risk, low probability but potentially recoverable patients. Thus, any discussion of such high risk patient populations needs to insure that there is a “outcome protection” for those forging these technologies and management strategies. 

Dick Shaw: This was a great discussion and leaves us with some answers but still many questions. [IAGS]