Current Advances in the Use of Therapeutic Hypothermia

During the 2019 Therapeutic Hypothermia and Temperature Management Symposium in Miami, several state-of-the-art lectures on current advances in the use of therapeutic hypothermia were presented. Dr. Justin Lundbye, Chief Medical Officer at The Greater Waterbury Health Network, moderated the sessions on various therapeutic windows for the use of hypothermia in preclinical studies as well as the use of prophylactic antibody treatment and other approaches used in clinical care. Dr. Neeraj Badjatia from the University of Maryland School of Medicine spoke on the implementation of targeted temperature management utilizing the latest evidence-based guidelines from the Neurocritical Care Society. Dr. Kees Polderman from the Anglia Ruskin School of Medicine in the United Kingdom reported on the beneficial effects of therapeutic hypothermia and targeted temperature management on myocardial function and myocardial electrophysiology. Finally, Dr. Patrick Lyden Department of Neurology, Cedars-Sinai Medical Center, provided an interesting discussion on an analysis of the Intravascular Cooling in the Treatment of Stroke 2 (ICTUS 2) trial utilizing targeted temperature management with thrombolytic therapy in acute stroke patients. Each of these presentations was well received and led to animated discussions among the attendees of the conference.

Dr. Kees H. Polderman: Pat, really interesting theory and it has a ring of truth, but a little pushback because the 24-hour time period for cooling does not totally come out of thin air. This is based on the processes that drive ischemia/reperfusion that take place over a period of ∼48 hours. In rats, we know that neuroexcitotoxicity is very important, it is a key thing and that is one that starts quickly and does not last that long. But in humans, other longer acting mechanisms such as neuroinflammation may be more important. We have no idea which mechanisms are key and which mechanisms are not, but if mechanisms such as neuroinflammation or apoptosis are more important (which might well be the case in nonrats), then we would need to extend rather than shorten the cooling time. So you might be right but it is not a certainty.

Dr. Patrick Lyden: There is no doubt that some of the things that go wrong in the brain go wrong over a period of time longer than 2 or 4 hours. The astrocytes are activated to protect the neurons, so when we cool the astrocytes we impede that repair process. The idea is that when we cool for 12 hours, we are impairing the brain's own ability to restore itself. I put a lot of data about astrocyte transference in that article. The theory would be: protect the neurons for 2 or 4 hours but then warm up and allow the astrocytes to start protecting the neurons and combating some of those processes that you are absolutely right do evolve over a longer period of time.

Question: Why not 1 hr?

Dr. Patrick Lyden: We only have data on 2 hours, but in the cell culture experiment, 1 hr was not as effective as 2 hours.

Question: Dr. Lyden, what I thought was really cool about the concept that you introduced with T1/T4 is it reminds me of a lot of what you see during hibernation with these intermittent arousal periods whether it be a bear or a squirrel wherein they are in a hypothermic temperature and they come out of it, and after 20 hours they go back into it and the process is repeated. It makes you think that there may be something important in this cycling that taps into this biochemistry. Same thing with the women's health initiative study that did not turn out so well whereby some think that giving estrogen in a chronic way is not as good as cycling estrogen for example, so maybe the concept is really interesting. Maybe you can comment on that.

Dr. Patrick Lyden: I did not know about that and I meant to find out more about the idea that bears are more relevant to us than squirrels because we need to do some field research.

Question: A long time ago we did some studies in transient global ischemia and we found out that if we cooled during the ischemic insult, it was pretty permanently protective, but if we started the hypothermia after the reperfusion injury and it was relatively brief, 2 hours, and we would get some nice protection at 3 or 7 days, but when we went out to do cognitive function, we lost that protection. So, Fred Colbourne and others decided to cool longer and felt that was why longer postischemic hypothermia was needed.

Dr. Patrick Lyden: That was your 2 vessel occlusion (2VO) model?

Comment: Yes.

Dr. Patrick Lyden: For those of you who do not know that model, the histopathological damage is seen in the hippocampus and forebrain which is the most glutamate rich part of the brain. It may very well be that cooling before an excitotoxic injury is really helpful, we just could not mimic that in a rodent model.

Question: A couple of times you alluded to in the graph with the intermittent cooling equating stroke and cardiac arrest and how you could apply the same principle to both. As we all know, there are some important differences. In stroke you have some injured brain and some not, you have the role of collateral flow, and in cardiac arrest you have a diffuse injury. I was wondering if you could explore a little bit the similarities you were drawing on and how the therapy might be different for cardiac arrest compared with stroke.

Dr. Patrick Lyden: The only reason why I said that is that graph is based on cell culture using the oxygen/glucose deprivation (OGD) model and the OGD model is really a model of cardiac arrest. It is a bizarre idea to culture neurons and then deprive them of oxygen and glucose. It is cell culture data. It is basic cell biology. What does it mean to a cardiac arrest model? We have not done the experiment. What does it mean in a stroke model? I did show you that 2 hours is better than 4 hours and we are looking for a way to replicate that in an in vivo cardiac arrest model.

Dr. Justin Lundbye: Kees, what is your thought on patients who have partial gene defects or something to that degree wherein they are a little more sensitive to prolonged QT, especially around drugs we may use, then have hypothermia on top of that. Even in a long QT syndrome, how do we handle those patients during hypothermia treatment?

Dr. Kees H. Polderman: First off, I have a tendency to use magnesium as an antishivering agent, and having a (very) high level of magnesium will be protective for that, and that is the first step. In general, nothing is ever one size fits all and that applies to temperature management and therapeutic cooling also. If the patient has additional risk for prolonged QT and maybe has atrial fibrillation, you might then use a slightly higher temperature, that is, 34°C (my standard temperature after cardiac arrest [CA] is 32°C). I try to avoid drugs that add to QT prolongation, but the literature does suggest that even in the presence of QT prolongation, the risk for arrhythmias is still low. The best evidence is probably from animal data, because there you can study each covariable and vary everything, and those studies would suggest the risk is not increased even in patients with prolonged QT.

Dr. Justin Lundbye: Have you ever paced someone at a higher rate just to see?

Dr. Kees H. Polderman: Well, some patients need a pacemaker even before cooling so yes. I would do that but I would pace them to a heart rate of 40 or 50 while undergoing hypothermia treatment, not 100. And I do not use dopamine for any reason in any patient.

Question: Excellent presentations. Are we still trying to find the right cycling because it was interesting that you found that 2 hours was beneficial, 4 hours was detrimental; with some review of the older literature about ischemic preconditioning and now trying to apply a benefit of postexposure, could it be that the issue was just the wrong frequency of the cycle?

Dr. Patrick Lyden: I am not quite sure whether I understand that question.

Question: Where there was similar literature before where you had a stuttering of exposure to ischemia that was somewhat protective but maybe a little less benefit with a similar postexposure duration of ischemia, perhaps the issue might also be with the hypothermia, is there an issue with finding the right cycle where 2 hours is relatively short compared with what we are used to seeing but that has not really been looked at in those time frames.

Dr. Patrick Lyden: Great thought. Preconditioning is a whole other issue but is the right number 2 or 4 or 6 hours? I only know the answer for cells in culture and we know that 4 hours does not work in a rat model but what is going to work in people? The right experiment would be an adaptive design wherein patients are randomized to 2 or 4 or 6 or 10 hours of hypothermia. Unfortunately, we can barely get funding to do a basic hypothermia trial. That key experiment will not be done, so we are going to have to somehow and I am hopeful that it might be a biomarker that will guide us on how long to titrate the first days of hypothermia.

Question: In the rare patients that you are concerned about them having adequate cardiac output during targeted temperature management (TTM), what markers do you use to figure that out? Let us say they are on renal replacement therapy so you do not have urine output, lactate tends to increase during hypothermia probably not related to oxidative metabolism. I find it tough at the bedside to figure out whether I have adequate cardiac output.

Dr. Kees H. Polderman: In general, it is not an easy question even under normothermic conditions. What I use is echocardiography and I use Picco catheters if there is a concern. I assess the patient clinically. I do look at lactate that does go up a little bit because of hypothermia, but it should not go crazy up. Using a combination of those assessments, that is what tells me if the patient is well perfused or not.

Question: I was really interested in this spot on temperature device. Can it be hooked to an iPhone^® on the side of a football field and actually measure temperature accurately?

Dr. Neeraj Badjatia: I have not really thought about the iPhone since I do not like them. I am an Android™ person so but, in general, I think it is a very simple device and I think we had had one of the device companies manufacture a cable for us and it took them less than a week so we could hook it up into their device. I do not think that will be a rate-limiting step.

Comment: We are very interested in athletes when they are very active and in hot weather what their core temperature is with the hypothesis being that maybe if they have a concussion during that time and they have a warm brain, they may do worse so we are struggling with a way to measure temperature.

Dr. Neeraj Badjatia: That is a great idea. I think with the concussed athlete, on the side of the field or at the time of their concussion, I would imagine this would be a great application for this because you will get an accurate measurement.

Dr. Justin Lundbye: Do all of you use prophylactic antibiotics for cardiac arrest patients? Pat, I think we have talked about this in the past but I could not remember whether you used them in any of your ICTUS trials.

Dr. Patrick Lyden: ICTUS avoided prophylactic antibiotics but had a very low threshold for aggressively treating once pneumonia was suspected.

Dr. Neeraj Badjatia: No, we do not use prophylactic antibiotics. I know there is some mixed literature on this topic. Some in Neurocritical Care have spoken about using it more routinely in the cardiac arrest patient population, but we do not do it.

Dr. Kees H. Polderman: It depends on the context of your baseline infection rate, so in the Netherlands, I used selective decontamination of the digestive tract, the infection rate is low there and what has been shown to reduce mortality, so every patient gets it not just the hypothermia patients, but in the United States it is not the case and I do not use prophylactic antibiotics, but most patients end up on antibiotics at some point anyway.

Dr. Neeraj Badjatia: Pat, can I ask you more about pneumonia and this is a definitional question. Clearly the 50% infection rate for the first study was not really real, it was a definition issue, getting down to 20% or 10% in your controls is more realistic, but we still struggle with how to define pneumonia and in our own weekly research conference. Question about how you define it, what objective criteria did you use, and on average when did your definition of pneumonia occur?

Dr. Patrick Lyden: So, that was my exact hypothesis. In ICTUS-L, we literally had an investigator at the bedside for 36 hours. Of course you are going to look at a chest X-ray and find something because you are bored to tears watching the Demerol drip so you are going to find stuff to treat, so I knew it had to be case ascertainment by us because it was not a blinded study. So, we adopted the Centers for Disease Control and Prevention (CDC) criteria that are very rigorous and very precise for when you can really say it is pneumonia and then those were adjudicated centrally and blindly, and if you noticed in the normothermia group the rate of infection was identical, it was 10% that disproved the idea that it could all be case ascertainment by us. It could not be because in a blinded way we had the exact same number in the normothermia group. In the ICTUS 2 study, all those pneumonias at 25% were modified white lung, positive cultures, cough, and the CDC criteria, and when we went back, we did not have the level of data for ICTUS-L but to the extent that we could when we went back, it really was not over calling white chest X-rays, these really were sick patients. I have no doubt that we dropped the true pneumonia rate between the two studies. It was not just a difference in definitions.

Dr. Neeraj Badjatia: When does pneumonia occur on average in your studies?

Dr. Patrick Lyden: I do not really remember but I think it was between 24 and 48 hours.

Dr. Neeraj Badjatia: To me as an intensivist, that is a different type of pneumonia. It sounds like you are dealing with an aspiration pneumonitis that may not be still in the end be technique related not necessarily temperature related and that technique of what you are doing with temperature is leading to impaired reflexes and more aspiration events and not necessarily pneumonia in the sense we think of in the intensive care unit.

Dr. Patrick Lyden: Absolutely, I agree.

Dr. Kees H. Polderman: Perhaps in ICTUS 3, we can use meperidine but at much lower levels than you did in ICTUS 1&2, if you use magnesium and counterwarming, you can reduce the sedatives that are needed and that is very important in an awake patient population because other awake patient trials in acute myocardial infarction (AMI) and stroke have sometimes had a slight increase in pneumonia risk, but not a doubling. It must be due to the fact that the patients are not intubated and their airway is not protected. I am thinking that meperidine at high doses could be a problem there.

Dr. Patrick Lyden: There was a lot of meperidine used in this study. We did do skin counterwarming in every patient and also gave buspirone. We did not require magnesium but sites were allowed to use magnesium. However, as evidenced by the respiratory rate, clearly they were over sedated patients at risk for aspiration.

Dr. Kees H. Polderman: If things were easy, everybody would be doing it.