During the 2018 Therapeutic Hypothermia and Temperature Management meeting in Miami, state-of-the-art lectures were presented on current advances in the use of therapeutic hypothermia in specific patient populations. For this round table discussion, experts in the field discussed studies using cooling strategies to treat ischemic stroke and acute myocardial infarction (AMI). Dr. Michael C. Kurz from the University of Alabama, Birmingham, moderated this interesting session. Participants included Dr. Patrick Lyden from Cedars-Sinai Medical Center, Los Angeles; Dr. Michael Dae from the University of California, San Francisco; and Dr. Marko Noc University Medical Center, Ljubljana, Slovenia. Dr. Lyden spoke on new studies in stroke research, including the use of a combination approach using selective cerebral hypothermia and mechanical artery recanalization. Dr. Michael Dae reported on past, current, and future studies in the area of therapeutic hypothermia for myocardial infarction (MI) summarizing current findings in this research area. Finally, Dr. Marko Noc presented the trial results from the Cool Acute Myocardial Infarction European pilot study which investigated strategies for reducing myocardial infarct size beyond primary percutaneous coronary intervention (PCI) using endovascular cooling before the myocardial reperfusion. This pilot trial justified a move forward to a larger pivotal trial. The presentations were all very informative and stimulated interesting discussions.
Question:
Why would therapeutic hypothermia only be effective for anterior circulation of the MI? Is this just related to the infarct size or something else?
Dr. Michael Dae: We think it is related to the size of the area at risk. So, with an left anterior descending coronary artery (LAD) occlusion, more myocardial mass is involved and if your end point is say a 25% reduction in infarct size using imaging methodology like magnetic resonance (MR) which has great resolution but still to measure changes, you need a big denominator. Also, I'll mention the fact that anteriors are the patients that have more heart failure, more death. They are the patients that really need help.
Dr. Marko Noc: I think this is the reason although I had a different hypothesis when we initially found this “anterior paradox” in CHILL-MI study led by David Erlinge from Lund, Sweden. My hypothesis came from every day observations on usually good collaterals to acutely occluded right coronary artery (RCA) as opposed to LAD in ST-segment elevation myocardial infarction (STEMI) patients. I hypothesized that good collaterals before reperfusion provide at least partially “warm” retrograde reperfusion which lead to a smaller reduction in infarct size after “cold” anterograde reperfusion. Acutely occluded LAD has usually very limited collaterals from left circumflex artery (LCX)/RCA, which leads to predominantly “cold” reperfusion when the artery is opened. Unfortunately, this attractive hypothesis was rejected by post hoc subgroup analyses of CHILL-MI and RAPID-MICE trials.
Question:
How about ultrasound upon admission before cooling is started so we can see if it is effective or not effective?
Dr. Marko Noc: Admission echocardiography in STEMI is very useful in patients with hemodynamic instability because it is important to understand the mechanism of hemodynamic deterioration. It takes only a couple of minutes to see if there is a left/right/global pump failure or maybe a mechanical defect behind. By adding immediate coronary angiography, the key information to tailor immediate treatment strategy is available. However, for the purpose of COOL AMI Pivotal study which enrolls only stable anterior STEMI, the role of immediate echocardiography is limited and it is not mandatory.
Question:
Dr. Lyden, for the two trials we keep seeing the pneumonia, are we cooling the vent circuit which could potentially lead to the pneumonia? For the COOL-MI people, I was wondering if you thought about doing the injection of chilled saline to the area as the same that was done in the
Intravascular Cooling in the Treatment of Stroke (ICTUS) trial?
Dr. Patrick Lyden: So the pneumonia problem occurs in ventilated and nonventilated patients. It occurs with and without evidence of aspiration. There is no doubt that stroke patients aspirate but stroke patients that are cooled get pneumonia more often. What we did that helped a lot was aspiration precautions. A lot of good nursing care really cut pneumonia risk in half but yet there is still a residual pneumonia rate, and a couple of things that might be at work is central immune suppression syndrome that we don't know a lot about but it is definitely true and that occurs in stroke but it also occurs in cooling. So to test that hypothesis, in ICTUS-2 we looked at C-reactive protein (CRP) to try and detect an effect on central immune suppression. We found that in fact the stroke patients who developed pneumonia were the ones that had the highest change scores in CRPs. CRP went from baseline to 24 hours later, fivefold higher in patients who developed pneumonia except in the cooling group. So, everyone who was cooled also had a change in their CRPs. The increase in CRP did predict pneumonia in normothermic patients but it predicted basically stress of the hypothermia which replicated the stress of the pneumonia. We are back to square 1.
Dr. Michael Dae: The idea of infusing cold saline in the cath lab has been tested in some preliminary studies in animals and I think it has been shown you can lower coronary blood temperature with infusing saline through a catheter and in fact I think I just found out a week ago that a trial is about to start to test that concept. I have a few concerns. I think the proposal is to do 10 minutes of hypothermia before reperfusion followed by 10 minutes of hypothermia after reperfusion then rewarm. I don't think that is enough time. There is ongoing injury after reperfusion. It may possibly be too brief. Also, I must admit there has only been one animal study that has looked at efficacy of cold saline infusion. This study was done in 2007, and the protocol was entirely different compared to what they are doing now. So, it has never been tested, even in animal models. I think the study is premature but you never know. Also, I am concerned about the possibility for arrhythmias. In another one of my lives, I studied sympathetic innervation to the heart using an imaging agent and was made aware of the concept that dispersion of repolarization of the heart is one of the major substrates for arrhythmias. So, think about this. You are doing localized cooling in a portion of the heart and the rest of the heart is still at normothermia. Localized cooling changes action potential durations so I think it is a setup for reperfusion arrhythmia but time will tell.
Dr. Marko Noc: I just wanted to add that the concept is very attractive because it avoids systemic hypothermia and everything becomes pure interventional procedure without the need for antishivering protocol and endovascular cooling catheter/machine. But I agree with Michael that systemic cooling and regional myocardial cooling are very different concepts, and infarct size reduction capability of regional cooling is still to be proven. However, if regional cooling indeed works, it will be a winner because it is much less complicated from a clinical point of view.
Question:
I think the cardiologists have really penetrated this issue the most. One thing I would like to ask you about is how do you handle the reperfusion period with regards to temperature control? I understand you cool for 3 hours then what is the natural trajectory for temperature in the next 24 hours? Do your patients get fever and if so, what do you do about it or do they not get fever?
Dr. Marko Noc: The University of Lund team with David Erlinge and Mathias Gotberg developed the cooling protocol which is now being used also in COOL AMI trials. They deserve a lot of credit for that. Mathias and David initially performed a small pilot study for 3 hours of postreperfusion cooling which was positive. In a larger CHILL-MI trial, postreperfusion cooling was, because of practical reasons, reduced to 1 hour and the study was negative. I am not saying this was the reason for a negative trial but I often question myself if additional 2 hours of postresuscitation cooling would make CHILL-MI a positive study. My hypothesis is that extended postreperfusion cooling might be important because in animal experiments, LAD is occluded by ligation or balloon and when released, reperfusion is immediate at epicardial, as well as on a microcirculatory, level. In STEMI patients, coronary occlusion is caused by thrombus, and primary PCI immediately restores epicardial patency but more gradual reperfusion is observed at a microcirculatory level because of distal plaque-thrombus embolization. With hypothermia extended beyond epicardial patency, gradual reperfusion at the microcirculatory tissue level is still “cold” and might further reduce infarct size. In CHILL-MI, we looked up a subgroup of patients with spontaneous reperfusion at the initial coronary angiography with obviously not much cooling before epicardial reperfusion. Hypothermia worked similarly as in the patients with completely occluded coronary artery indicating possible benefit of cooling during microvascular reperfusion. I think we are now slowly starting to realize.
Question:
But then what about temperature in the next 24 hours? What is the patient's natural trajectory of temperature?
Dr. Marko Noc: If you take a regular STEMI without hypothermia, CRP goes even beyond 100 mg/L, white blood cells increase beyond 10, and temperature goes up even above 38°C in early postinfarction period. This represents a physiological response to myocardial necrosis. We published on that >15 years ago. In our COOL AMI patients, gradual rewarming is started after 3 hours. After reaching normothermia, these patients may still develop similar temperature and inflammatory response as regular STEMIs during subsequent days.
Question:
Do you monitor temperature?
Dr. Marko Noc: We monitor continuously during the cooling/rewarming and hourly during the subsequent days.
Question:
This is so interesting. This is what is so good with a meeting of this kind because we can learn from each other. It may be the same disease and we should join all these data together and do the perfect trial in the near future.
Dr. Marko Noc: I always challenge myself if the brain really is so different from the heart? Maybe what we are doing in STEMI is what should be done in stroke and out-of-hospital cardiac arrest (OHCA) before achieving return of spontaneous circulation (ROSC). The global idea is to start cooling as soon as possible after the onset of ischemic insult and maintain it during reperfusion and postreperfusion period for certain time which probably depends upon the organ and severity of ischemic insult.
Question:
We have asked that question at this meeting, the duration of cooling after reperfusion in MI, and I think that the response has been the inflammatory response is different and therefore you don't have to have that long of a cooling period to target that secondary inflammatory response. It has always been puzzling to me and now that you have talked about some of your own data that has to be something in the back of your heads. Hopefully, this trial will be positive. Maybe as you get more severe MI where you may have to do something more additive, maybe then the duration of hypothermia might be something to consider and slow rewarming and all the things that we have talked about.
Dr. Patrick Lyden: I think one thing that is absolutely clear is that one duration isn't the answer. So, we set 24 hours for one study and 2 hours for another and 6 hours for another study. This way we completely ignore the patient, and the patient has had a different duration of downtime whether it be stroke or cardiac arrest or AMI, a different reperfusion time and then a different interval to target temperature. All of these variables should be considered in deriving a duration based on how long the patient was ischemic and when they were reperfused.
Dr. Michael C. Kurz: Dr. Lyden has a great point. While we wish to personalize duration of cooling, we have few biological markers or parameters to guide us. Collectively resuscitation scientists discuss at length which temperature and duration of rewarming are correct, but there are many other parameters regarding temperature management that we haven't examined or don't discuss which may have more impact on the injury involved.
Dr. Patrick Lyden: I can tell you there are at least two approaches. One approach is that the duration of cooling is some numerical multiple of the duration of cooling time. Dr. Fred Colbourne has shown this in the hippocampal ischemia model that the longer the duration of cooling or the longer to target, the longer you have to cool. So you cool longer to get the benefit. The other approach is a biomarker. If you could find the magic indicator then you could titrate your duration of cooling based on that biomarker.
Question:
Why haven't we done animal studies looking at longer durations of cooling? Probably the short duration works but we don't know if longer duration might work better. To my knowledge, none of the animal studies has tried this.
Dr. Michael Dae: You are absolutely right. The study that we did a number of years ago where we cooled, reperfused, and put the machine to rewarm, and it took about 3 hours just to get back to normothermia; we said that is what we will do in the pilot study. The 3 hours worked and we stuck with it. I will tell you that there is a trial underway now in Brazil that is comparing 1 hour of hypothermia to 3 hours of hypothermia. I think over the next year, we will have some data if it makes a difference.
Question:
I am thinking more like 24 hours. So, you might get more benefit that way. Right now we have no idea. So, this would be hard to do in awake humans but it certainly is possible if we get a signal from animal data that longer cooling works better then we could consider doing such a trial.
Dr. Michael Dae: It should be looked at but I would ask Dr. Noc. I know interventional cardiologists are very impatient people and anytime you add an extra 5 minutes to anything they have an issue.
Dr. Marko Noc: You are right. Even those 18 minutes of waiting in the COOL AMI study is long for normal interventional cardiologist without sufficient scientific background on possible benefits of cooling before reperfusion. You have to entertain him to achieve that he will not push the wire too soon. However, with COOL AMI Pilot study showing 7% absolute infarct size reduction despite 17 minutes of cooling related delay in reperfusion is now easier.
Question:
I had a logistic question for Dr. Noc. It sounds like in your intervention arm there was a lot going on in a short period of time. Do you have extra staff? Did you get a sense that there was an extra clinical burden to that extra work that might be important?
Dr. Marko Noc: In our hospital we have a critical care unit (CCU) physician joining interventional cardiologist in the cath lab for each STEMI, non-STEMI, and OHCA. Interventional cardiologist can really focus only on PCI, and the rest of the work is done by the CCU physician. This concept helps a lot to this study because you need additional people for cooling, antishivering protocol, and case report form (CRF)-related administrative work. Despite that, additional extra physician is usually coming from home to help. It is easier with more people but you have to have a “conductor” who says now do this, now push the button, don't push the wire… and takes care of the clock.
Dr. Michael C. Kurz: Dr. Noc, in light of your investigation, how do you, or other members of the panel, handle the bureaucratic quality reporting for STEMI and other time sensitive disease states? We have not discussed what panelists believe will convince our colleagues to evaluate if temperature management works and intervention can be delayed past usually acceptable windows.
Dr. Marko Noc: Indeed, what about extension in door to balloon delay which will, in turn, increase the infarct size? A very legitimate question, but now we have, as I already pointed out, the results of COOL AMI Pilot showing a 7% absolute reduction in infarct size despite 17 minutes of cooling-related delay in reperfusion. This is a very strong argument which justifies the study protocol and shows that we are not doing any harm. Rather opposite.
Dr. Patrick Lyden: So time is brain, time is muscle and I think the burden is on us to show a clear and persuasive benefit to any delay whether it is door to lytic, door to needle, door to balloon, if we are going to cause a delay for AMI or acute ischemic stroke (AIS) then we need to show a very powerful benefit in exchange for that. It might be that the better approach is rather than trying to figure out the logistics and make people fold their arms for 15 minutes is to find more powerful catheters so that the temperature target is reached more quickly. Also, we could consider the use of intravascular cooling whether it is intracerebral or intracoronary cooling as a bridge to that powerful catheter getting the core temperature down. Because really telling people to wait or pause is a losing strategy.
Comment: We are in the process of recruiting U.S. sites for a follow on trial to what Marko has shown and some people are clearly not interested. They are very concerned about benchmarking but others when they see the science which Michael showed and the work that Marko showed, they are receptive to the idea that a relatively brief delay from door to first medical contact to PCI is perhaps associated with a reduction in infarct size and also the sponsor is working with the sites to make sure there is minimal deleterious impact on door to PCI time as possible. It is a work in progress. Door to PCI is a clear focus but it may be in the future that door to target temperature becomes the goal. That is pretty speculative but I don't think we are there yet.
Dr. Michael Dae: I agree and speaking of speculation, I would like to see a billboard in every metropolitan area that says don't reperfuse faster, reperfuse better.
Comment: If we provide proof that this works, then we will probably move to cooling in the ambulance so as not to lose time for reperfusion, the patient will already arrive hypothermic in the emergency department and cooling continues with different ways to the cath lab so you lose less time in the cath lab. Then we know that this works.
Dr. Marko Noc: This is exactly what I am thinking and it is at the moment rather futuristic. Let us start with electrocardiogram-confirmed STEMI on the field who gets Aspirin and heparin already at the first medical contact. Maybe in the future we will have an intravenous agent which would bring setup point in hypothalamus down to 32°C and induce pharmacological hypothermia already on the way to the hospital. No shivering, no need for device cooling, just spontaneous cooling with vasodilation and sweating. Once arriving to the cath lab, cooling will be augmented by endovascular catheter implanted before reperfusion as we do now. Such cooling would cover not only reperfusion and early postreperfusion interval but also ischemic time on the way to the hospital which is likely to further reduce infarct size.
Dr. Michael Dae: I agree. Another difference in the animal studies from the human studies is that almost all of the animal studies provide hypothermia throughout most of the ischemic period and it is just not possible to do that in patients. The typical ischemic time in patients is 4–6 hours so if we are good with the time of first medical contact and start inducing hypothermia, I think that is going to be a home run. That is a technical challenge but that is an opportunity for those who are still designing devices and figuring out new things to do. We still need a successful way to induce hypothermia in the field effectively and surface cooling is probably not that way because it is too slow.
Question:
The first point I would say is that keeping these people cold for 3 hours is a bit of a pain and to elongate that for a minimal gain or a gain we don't know would be disappointing. I wonder if we do have a positive trial in the pivotal study whether people will do it according to the study protocol or actually they will get them to a target temperature, they will deliver them the reperfusion therapy at say 32.2°C with a temperature directed reperfusion rather than a time directed reperfusion and actually afterward there may be some leniency on how you do the 3 hours afterward because it does have implications for your nursing staff, for your monitoring, for your ambulation, for your eating, for your drinking, and for your sedation and all of the other things that we know which can be a challenge with conscious sedation, that will be interesting to see what happens. Dr. Lyden, in the EURO-COOL hypothermia trial that is being planned by the Pills group which targets local hypothermia in coronaries in 200 patient group, they are occluding the artery so that they are infusing saline down the balloon while it is occluding. The data you suggested seemed in the brain that reperfusion could already have occurred, is that right? Or are people thinking about in the brain putting a balloon up and infusing like coronaries and keeping it so there is no flow and then the first bit of reperfusion is cold, do you think that is less important?
Dr. Patrick Lyden: I personally think that is less important. We did a study where we got our rats to target temperature before reperfusion and we found no hint of benefit compared to postreperfusion cooling. So, I don't know if the brain is that different from the heart in that respect but I don't imagine anybody is going to blow up a balloon and avoid reperfusion. They will cross the clot with a microcatheter and inject saline while they are navigating their thrombectomy device but I don't see any neurointerventionalist blowing up a balloon to delay reperfusion.
Question:
To my knowledge, the fastest means of cooling is the immersive cooling device that gets people super cold, super quick, dangerously quick which doesn't require administration of ice cold saline and can be done relatively quickly without adding a delay to getting a cath. I think this is made by Life Recovery Systems. Has that been looked at?
Comment: As far as I know just in rabbits.
Comment: They tried to initiate those studies in humans but I don't think they have had success.
Comment: You would have to intubate the patient. The device is like a kiddy pool.
Comment: They have tested that device in humans in Vietnam and it was in the emergency room and the cooling rates were not as good as expected. I think 3 degrees per hour in that context. If we find a benefit and we move to prehospital cooling, we should resurrect the peritoneal cooling system because that is suitable for prehospital cooling. It is not that easy but if the benefit is shown then I think that would be a useful tool for that. This procedure basically flushes cold saline into the peritoneal cavity.
Question:
I have a question for Dr. Lyden. Thrombectomy may open up the therapeutic window for reperfusion. Of course we are concerned about
tissue Plasminogen Activator (tPA) if you give it late because it may convert a pale infarct to a hemorrhagic infarct. That is not the case with mechanical removal of a clot?
Dr. Patrick Lyden: Those late thrombectomy trials did not have an increase in hemorrhage primarily because they used imaging to carefully select those patients that have penumbra. As soon as the core infarct gets above 70 mL then you theoretically increase the risk of hemorrhages and certainly above 100 mL you don't reperfuse because you definitely increase the risk of hemorrhage.
