Transcript Video Temperature Control in ICM Traumatic Brain Injury Back to Symposium Dr. Coles discusses Traumatic Brain Injury in ICM Good afternoon to everybody and welcome to this webinar which is focused on the management of a temperature after traumatic brain injury. Thank you for being with us this afternoon. This is actually the 3rd webinar of the series which is focused on the control of temperature in different pathologies in the ICU. There was a previous webinar on cardiac arrest, one on cerebrovascular disease, and for those, I would like to thank BD for the support and the ICM for the organization and the support. Today, as mentioned, we will focus on traumatic brain injury where the control of the temperature is fundamental. And to discuss this, we have two incredible speakers, two very expert speakers in the field. We have Dr. Andrea Lavigno from the University of Cambridge and Professor Jonathan Cole still from the University of Cambridge. And now I will leave the floor first to Andrea and then to Jonathan. And I really would like to to thank you for the possibility to introduce Andrea Lavigno, who is a friend and colleague. We have worked together a lot on consensus recommendations regarding temperature management in and cerebrovascular disease. He's a consultant in Cambridge, UK. And uh he will start with a talk on uh uh temperature control and precision medicine. Andrea, the floor is yours and thank you very much. Thank you very much, Kiara, and uh good afternoon. This is a page with my declaration of interest and this uh uh webinar series has been supported uh by BD. And uh this is the culmination of a long arching project that sees temperature as a key aspect of high quality medicine, in a variety of settings including um, Post-cardiac arrest care and there was a recent update only a few weeks ago with uh again the recommendation that temperature control should be maintained in comatose patients, the changes that the duration of treatment should be at least 36 hours. In the field of uh um uh vascular brain injury that includes uh um stroke, subarachnoid hemorrhage, and uh um intraparenchymal hemorrhage, the, the, the, the big update is the publishing of the Intrepid trial, which compared the fever prevention with standard care. Showing that uh uh fever prevention does to some extent affect temperature, but they, they, they were unable to detect any difference in outcome. The, the authors note that 25% of patients never develop fever, even if these patients were patients that required the ICU for a prolonged period of time and so considered to be at higher risk. Uh, and that fits with the, uh, our own or recommendations uh uh for temperature management following vascular brain injury, which includes subarachnoid MRHA, stroke, and ICH. And the pathway towards secondary neurological injury is known to be affected by ischemia, edema, inflammation, and as is the case with traumatic brain injury by raised intracranial pressure, and we know that temperature modulation, Affects all of these aspects uh uh very significantly, and the um, The key recommendation in this setting was that we need to have a deliberate strategy for temperature control in these patients, which is based on continuous monitoring, which is the only way to detect promptly uh fever in vulnerable patients and, and to treat it effectively. And thinking about temperature in first principles, temperature measures the average kinetic energy of particles within a medium, and we define the centigrade scale, where energy is so little that water solidifying twice and the boiling 0.100 degrees when the energy is so high that molecules of water escape into the form of vapor. And, and we know that in a healthy human temperature is controlled within a very, very narrow range, and we do so uh with a very high expenditure of energy, uh, and we have uh uh. Neuroendocrine um and metabolic modulation that can control that, uh, like sweating, vasodilation, uh, or shivering, when we want to increase temperature, but a key to it are behavioral responses, uh, which is uh sort of uh getting a, a, a warm coat or staying inside in winter or seeking, Uh, uh, cold shelter or air conditioned spaces, uh, in the heat of summer and all of those behavioral responses are, are very severely blunted in a patient that's, uh, um, undergoing a general anesthetic or sedated in the ICU. So there is a combination of, uh, um, thermal dysregulation. Uh, in, in that context. And we know how profound a small change in temperature is uh uh is when we experience uh a fever, uh we feel immediately ill, and but, and, and uh. Obviously one of the commonest medical intervention is to try and control fever, but fever itself is part of the innate and of the adaptive response against pathogens and particularly effective against the viruses, but also in in sepsis we can see uh this is a relatively recent large Swedish court. How being able to mount a fever response in response to sepsis is very strongly correlated with outcome, and you see a mortality rate going from 40 to 50% in patients that have hypothermic or normothermic. Uh sepsis compared to mortality that's below 15 to 10% in patients that are able to mount, um, a, a strong fever response. However, in the context of neuroprotection, we think of fever as being. Potentially very problematic in this graph is at the center, a normal physiological state where you have oxygen delivery to the brain, which is cerebral blood flow multiplied by the content of oxygen of blood, is appropriate in respect to the cerebral metabolic rate of oxygen, so consumption. Is matched by oxygen delivery, but cerebral blood flow, or or or oxygen delivery to the brain can be impaired by a variety of uh uh of scenarios. Obviously when we have hypotension or hypoxia, the, the, the, the key response to of the intensivity is to correct, uh hypoxia or or hypotension by putting in place, you know, securing the air or ventilating the patient or providing, vasopressor or resuscitation, but in other cases, for example, when you have ICP or vasospasm, and you have a critical reduction in oxygen delivery without a matched reduction in oxygen demand, you experience ischemia, and so what we can do in these cases is the, To induce uh uh sedation or hypothermia, to reduce uh metabolic rate, and contrary to that, if uh uh fever or seizure occur, we would accelerate the ischemic insult. So really what we are trying to deliver by sedating and controlling temperature with deliberate intent in these patients, the brain at risk is to provide a state of coupled hyperperfusion and hypermetabolism, which perhaps can give time for swelling or or or or or a space occupying lesion to be evacuated so that normal physiology can restore whilst minimizing. Neuronal loss and uh in this context, uh, we do, uh, and have recommended, uh, controlled normothermia and mild hypothermia and a tiered approach, uh uh in our recent recommendations that John will be discussing at length. As a preparatory work to that publication, we did a systematic review with an attempt to uh provide a meta-analytical basis for this. And those familiar with this field of research will know that there is enough evidence that can be meta-analytically. Uh, analyzed, it's very heterogeneous, uh, um, cohort of patients with different study design. So the question and the debate that's quite recurrent is, are those recommendations evidence-based? Is this evidence-based medicine? And now evidence-based medicine was a key revolution in modern medicine that's now almost 3 decades old, and it means the combination of the best evidence. With the physician expertise to provide patient centered care, what it is not, it's the oversimplification that says treatment is effective or ineffective or or or or beneficial in a condition. Because very very few situations are simplified like that, and certainly not in an information rich environment such as the ICU. Uh, one of my main bugbears is, is uh steroids are harmful in traumatic brain injury, uh, which unfortunately we won't have time to discuss, but the, this oversimplification of what evidence-based medicine might be or might not be. Is exemplified with this now classic. Uh, uh, publication on the BMJ, which, uh, uh, have reflected on the fact that parachutes may reduce the risk of injury, uh, after gravitational challenges, but this is not evidence-based because there is no randomized trial to demonstrate that. And similarly, what evidence-based medicine is not is a color by number attitude and or exercise. It still requires clinical judgment and understanding of the complex dynamics and physiological or pathophysiological interaction of the patients in front of us. One of the best examples in the context of neuroprotection. It is the discussion about what the critical ICP value might be, and by seeing patients doing infusion studies with communicating hydrocephalus and being able to talk with ICP in in excess of 40 or 50 millimeters of mercury. Realize that there isn't such a thing as a critical value of ICP. So for the case on the left of your screen where you can see the the the sulci are visible everywhere, the basal cisterns are open, that patient is likely to be able to tolerate elevated ICP if your CPP is maintained. On the right side of the screen, even small elevation of ICP above atmospheric pressure are going to cause pressure gradients and herniation, so we would be very, very aggressive in controlling ICP in the patient on the right, whereas we might remove a bolt on the patient on the, on the left if we're trying to extubate them. So the point of that is that the application of evidence-based principles should be applied to the ICU, but we should tailor what we do and the intensity of our treatments to the individual and their response to intervention, and we have endpoints such as ICP, brain oxygenation. Uh, and, and, and other endpoints that can inform, uh, um, the intensity and how aggressively we control temperature. and so on, on a very simplified, um, uh, kind of uh approach, you'd say that temperature selection, if you have a patient on the right that doesn't have an acute. injury or affecting their brain or spine, you might want to tolerate higher temperature, especially in the context of sepsis, because that is effectively protective. In patients who have a compensated acute brain injury, you would want to keep them in the controlled nothermia range. And like our CT scan on the right, if you have a patient that are at immediate risk of herniation or where you have markers of compromised brain perfusion, you may want to kind of move them to that area to the left of that graph where you have coupled hypoperfusion, hypo hypometabolism. And you determine the duration of this uh by uh reassessing uh uh imaging, um, neuromonitoring, uh, and also the eventual side effects that you might experience. So, thank you very much for following this introduction, and it is my absolute pleasure to introduce our friend and colleague, uh, uh, Professor Jonathan Coles who had a long-standing interest in uh in uh um in uh neuroimaging and clinical neurosciences. He's our clinical professor of intensive care medicine here at Cambridge. He's a colleague on NCCU. Uh, he's also the president of the neuroanesthesia and uh Critical Care Society, and, uh, he has, uh, uh, an important role as a lead examiner for FICAM, which is our faculty of intensive care medicine, and he, uh, has a long record of very impactful publications on the field. So thank you very much, and John, the floor is yours. Thank you for that introduction, Andrea, that's very kind of you and as Andrea says, it falls to me um to give the talk um centered around temperature control er within traumatic brain injury. And first of all, I'd just like to disclose that over the years I've received financial support for my academic research from a variety of different research charities and organizations. And also as Andrea mentions, I'm currently employed by the University of Cambridge. And I'm currently also the president of the neuro anesthesia and Critical Care Society within the UK. So thank you for the opportunity to talk to you. First of all, um, I think it's worth highlighting that we all know that, uh, severe traumatic brain injury remains of significant public health concern for us. And if we just look just within patients within intensive care, we see that mortality is still over 20% in a recent publication from the Centre TBI Study. Over 40% of patients are unfortunately die or are disabled, um, uh, following trauma to the brain, and over 80% make a less than complete recovery, so this is a substantial burden on society. And although we've had um updated guidance from um the Brain Trauma Foundation about how to manage patients with severe traumatic brain injury, these stop short or fall short of providing detailed, um, ways to manage a patient at the bedside and and instead emphasize that the lack of evidence to support any level one recommendation for the use of any intervention that could modulate, Intracranial pressure or cerebral perfusion pressure in this context. Instead, they mentioned that ICP and CPPPP monitoring is recommended because it's associated with decreased to weak mortality, and that targets based on population averages suggest that patients with an ICP of greater than 22 should retrieve treatment because they're more likely to die. And but also that we should target cerebral fusion pressure values between about 60 to 70 millimeters of the mercury because that's associated with increased survival and a more likely incidence of favorable outcome. But we know from our own studies, and this is work from myself and others in Cambridge, where we've used physiological imaging, PET imaging using triple oxygen PET to demonstrate the extent of individual heterogeneity in physiology, not only between patients but within different areas of the brain within the same patient, and that this heterogeneity changes from day to day. So it's a very complex pattern of physiological changes that we see. These data show evidence of classical ischemia based on a high oxygen extraction fraction, more than 75%, occurring within some regions of patients, particularly within the 1st 24 hours shown here in red, but also extending out to about 2 weeks. Injury, but the important point for this deranged physiology is that you see it despite maintaining cerebral perfusion pressure above that average population value of 60 millimeters of mercury, and it also often occurs in the absence of any rise in intracranial pressure. So why should we bother with temperature control and traumatic brain injury? Well, we've heard nicely from the introduction from Andrea that hypothermia is associated with poor outcome in patients with traumatic brain injury because it increases metabolic demand and therefore puts the brain at greater risk of ischemic injury. It lowers the seizure threshold. It's also With an increased risk of secondary injuries arising from excito toxicity, blood-brain barrier permeability, or increased inflammation in a disordered fashion, and this can result in calcium influx, apoptosis, or cell death, ultimately leading to cerebral atrophy and poor functional outcome for patients. Moderate hypothermia on the other hand, can be neuroprotective because it can decrease the demand for oxygenated blood and therefore lessen the chance of ischemia in vulnerable brains and decrease the secondary injury processes that we've already discussed. Um, and look back at the basic physiology, Andrea mentioned that cerebral oxygen metabolism in health is normally tightly coupled to cerebral blood flow, and this enables the delivery of oxygenated blood to areas of increased metabolic demand. That increased blood flow is due to vasodilatation, which means an increase in cerebral blood volume, which in health can be accommodated, but in patients with raised intracranial pressure following traumatic brain injury, that can increase intracranial pressure still further. So to guard against this, we can take advantage of this coupling, using metabolic suppressants like hypothermia to reduce cerebral blood flow. Due to vasoconstriction, reducing salable blood flow volume and increasing our ability to control intracranial pressure in these patients. And we can use this to lower basal metabolism, to to lower cerebral metabolism, to basal levels required for sustaining cellular integrity, and we can monitor how much dose of intravenous anesthetic agent or hypothermia we use by monitoring electrical responses. at the bedside using processed EEG or full EEG to ensure that we reach a pattern of birth suppression and don't expose patients to the adverse consequences of these treatments which, as we know from hypothermia can lead to an increased susceptibility to infection, coagulopathy, and arrhythmia or hypotension. So how should temperature control fit into our treatment protocols for these patients? Well, Cambridge, like other centers treating these patients, has a treatment protocol in a staged fashion of escalating therapy intensity, and we reserve treatments with higher risk of side effects such as hypothermia. Um, deep sedation or barbiturate coma or decompressive craniectomy to the later stages of these treatment protocols to ensure that we don't expose patients to the higher risks associated with these treatments and only titrate them to those that need them. These algorithms in local centers are based on international consensus, and these are the, these are the Seattle International Severe Traumatic Brain Injury Consensus conference guidelines, the civic guidelines, which were published in 2019 in intensive care medicine. And they, as discussed, provide a tiered management approach, but the basic principle being we should use the lowest tier possible to manage intracranial pressure and cerebral perfusion pressure. These guidelines reserve mild hypothermia, that is a temperature between 35 to 36 °C, using active cooling measures as a tier-free therapy used alongside barbiturate coma or secondary decompressive craniectomy. And why is this? Well, if we look at recent evidence that has looked at all the studies, and, and there's something that Andrea alluded to, um, this systematic review and meta-analysis showed that the heterogeneity of the studies concerned, and although it split them up into those that were of low risk of bias compared to high risk of bias. The suggestion of improved outcome from hypothermia was limited to those studies with a potential high risk of bias. So overall, based on the current studies with a low risk of bias, there's no evidence of any mortality or functional outcome benefit from hypothermia in the management of traumatic brain injury. And although you might think that's an answer to the question, I think if we look at these studies, and if you just focus on the large studies of hypothermia, the Eurofm study, which showed evidence of harm from hypothermia, and the ANZACs Polar study which showed no benefit from hypothermia. If you look at a more detail, in both cases they were hypothermia was used as an early intervention and used universally and early in the polar study, um, rather than using it at a time when the other measures that we use for controlling intracranial pressure had been exhausted or had failed to control. And if we borrow a line from Shakespeare's Hamlet here, we can see that these desperate diseases grown by desperate appliances are released. What that translates into, we should be using more desperate measures when the times are more desperate, when we've exhausted the ability to control intracranial pressure and maintain cerebral perfusion pressure using the lower tier therapy. That's only when we should reach for these therapies such as hypothermia and in that instance, the risk benefit ratio may be more favorable. And what we're talking about here is patients in extremis. This is a patient you can see in the top left who's had a um a traumatic brain injury, they've got a large extradural hematoma, which is causing substantial mass effect and imminent danger of herniating and death, like the postmortem specimen we can see on the right. And in this circumstance, the only thing that's going to save this patient's life is taking them to theater and evacuating that clot. But whilst they're being prepared for theater, you may consider some other life saving measures. These include, as an example, hyperventilation to lower the arterial partial pressure of carbon dioxide, to reduce cerebral blood volume, and to reduce intracranial pressure as a temporizing measure. You could also consider administering chilled or ice cold fluids to lower the temperature rapidly and act as a potential neuroprotective strategy. And this is a different scenario to the one you see below in a patient who's already had a craniectomy. The subdural hematoma that you see has already been partially evacuated, and the intracranial pressure is controlled to the high teens or around 20 millimeters of mercury. But when we look in this patient, we may still find evidence of a reduced cerebral blood flow. Potentially associated with reduced oxygen delivery and a higher risk of further ischemic injury, but in that instance, we might look to less invasive treatments trying to optimize cerebral perfusion and cerebral oxygen delivery. So we need to, in these different scenarios, we need to calibrate our intensive care treatment to ensure that we maximize the benefits without exposing patients to higher risk. So if we go back and look at the the the existing studies of hypothermia in a little more detail, first we'll look at the Eurofm study published in the New England Journal of Medicine in 2015. And as we know it was an excellent study, a large study enrolling 387 patients, but the headline finding was that favorable outcome occurred less frequently in the patients in the hypothermia group compared to the group that received standard care. And so the conclusion was that hypothermia was no better than standard care alone. If we look at the protocol, patients with traumatic brain injury were recruited, um, and, and at recruitment they were receiving stage one treatment as defined by sedation, ventilation, and intravenous fluids to maintain a mean arterial pressure of 80 millimeters of mercury or more. And they were randomized once ICP monitoring had been commenced, and ICP was found to be higher than 20 millimeters of mercury for at least 5 minutes in patients only receiving stage one treatments. Once randomized, those that were placed within the hypothermia group received targeted temperature or titrated temperature control to 32 to 35 °C. And could receive further stage 2 treatments if ICP still remained elevated, and in the control group, standard treatment was not prescribed but included osmotherapy. And ionotropes to maintain a cerebral fusion pressure of at least 160 millimeters of mercury. Further options for both groups included barbiturate coma or decompressive craniectomy if ICP was still difficult to control. Outcome was day 28, hospital discharge or or death, and 6 month follow-up using the Glasgow outcome score. And if we look at the study results, clearly the headline finding was that those in the hypothermia group showed lower survival, um, and adverse outcome. But if we look at the early physiology, we can see that there's very little or no difference at all in terms of intracranial pressure and cerebral perfusion pressure, and the bottom line here is that these patients' studies did not have what we would define as more desperate disease, and the intervention did not assess the impact of hypothermia in patients with intracranial hypertension that was refractory to stage 2 treatments. The other study that we'll briefly mention is the ANZAC Polar study, which specifically was an assessment of early prophylactic hypothermia in the long-term outcome of patients with severe traumatic brain injury. So again, it's a large, excellent, well conducted study that looked at prophylactic hypothermia compared to standard treatment, and we can see that there was no difference in the rate of favorable outcome between the groups. But you can see here from this study, this was a prophylactic early intervention. And again, patients studied did not have refractory high ICP that had failed to manage with other less invasive treatments. Intracranial pressure over 20 millimeters of mercury was only seen at around 40% of monitoring period, and in both groups, the rate of decompression craniectomy was low, suggested that these patients did not have high intracranial pressure. So whilst these two studies are both um very good studies and address the question of whether the hypothermia should be used as an early treatment for traumatic brain injury, they do not assess whether it could be used as a later treatment in those that had failed initial measures to control intracranial pressure. And if we can consider this in line with other treatments for severe organ failure such as ECMO, which is now an accepted treatment option for severe respiratory failure because it treats extreme physiology when, Conventional therapy has already failed to to reverse the pathology. It's not used as a treatment for a specific diagnosis such as a diagnosis or label of influenza, or for less severe episodes of respiratory failure, or even for preventing. Severe respiratory failure, and we should consider this same context for hypothermia. We can conclude that there's no benefit from early hypothermia in patients with traumatic brain injury based on the findings of polar and Eurotherm. However, we do not know whether there is a role in cases of refractory high intracranial pressure, and to answer this question, we would need another relevant trial. So what about less desperate disease? Well, even in these, the consensus guidelines published by CIIC would suggest that we need to do more than use population averages for cerebral perfusion pressure, because otherwise if we don't use optimized cerebral perfusion pressure for individual patients, we may undertreat some patients and overtreat other patients. Therefore, not really benefiting either, and also to understand the dose and impact of a rise in intracranial pressure, we need to understand more about the downstream effects and therefore use our treatment and titrate our treatment to optimize physiology in these patients. Otherwise, as I say, we run the difference of providing a treatment that fits no one and doesn't benefit any of our patients with traumatic brain injury. We can characterize the dose, and this is excellent, very helpful data from Gert Meiffereut and the group of Leuven, where they have helped define ways of looking at the intensity of the intracranial pressure in salt. Not only on the level of intracranial pressure, but the duration of insult and look to see what impact that has on patient outcome. These graphs show in red patients with poor outcome compared to the darker colors in those who have had a good outcome in a large cohort of some 261 patients managed on the intensive care unit. And whilst you can see that patients tolerate poorly a rise in ICP above 25 to 30 for only a very short period of time leading to poor outcome. You can see that even mild increases of intracranial pressure in the region of about 15 millimeters of mercury can be harmful if they occur for a prolonged period of time, and importantly, the tolerance, tolerance, sorry, to these, these insults is reduced if cerebral perfusion pressure is below the lower limit of auto auto regulation. To describe that a bit in more detail, we can see from this uh paper now published some time ago from Lucia Steiner and the Cambridge Group, um, who used the pulsory activity index, a continuous measure of the relationship between the arterial pressure wave form and the intracranial pressure wave form. You can see, um, the range of cerebral fusion pressure at which auto regulation remains intact. A negative PRX suggests maintained autoregulation, whereas a positive PRX suggests that it is impaired, and you can see if you plot the data for individual patients over time, you can see a range of cerebral perfusion pressure where cerebral, which is optimal in terms of maintaining normal cerebral auto regulation. And then this paper, um, Lucius, uh Riel showed that optimal CPP varied between about 50 to 110 millimeters of mercury. And importantly, the management of patients close to this optimal perfusion pressure was associated with better outcomes. And so optimal cerebral perfusion pressure is a possible target for individualizing cerebral perfusion pressure at the bedside, and the cogitate group have looked at this, although it does require a further larger study to prove it in a prospective cohort of patients. So if we're looking at approaches to refine our treatment and our ICP management of patients with traumatic brain injury, clearly to minimize side effects we need to individualize therapy and target treatment. We should optimize steroid regulation. But we also could look to downstream events as a marker, and there are a variety of different markers that we could look at, but some of the important ones that we currently use are the partial pressure of oxygen within the brain with focal monitoring probes, or looking at brain metabolism and the impact of what our treatments have on that, both substrate delivery such as glucose, but also looking at indices of metabolic compromise like lactate and the lactate per uvo ratio. And clearly what we know is so blood flow within the brain is driven by the cerebral perfusion pressure, and when this is lowered in patients with high intracranial pressure, we can see a reduction in oxygen delivery resulting in disordered metabolism, sometimes ischemic patterns, but sometimes in other patterns of disordered metabolism associated. With poor outcome and we can monitor that at the bedside using a combination of not only intracranial pressure monitoring but monitoring of brain tissue oxygen and the adequacy of oxygen delivery to the brain, but also the adequacy of brain metabolism and the adequacy of substrate delivery like glucose to the injured brain. And we use this in our, in our hands, in our unit, but though we may initially start with a global target for cerebral fusion pressure of 60 to 70 millimeters of mercury, we'll use all of this monitoring information to individualize the treatment that we find our patients and individualize the CPP targets that we use and potentially individualize when and how we use control of temperature in the injured brain. And using these monitors and the available literature, you can develop a range of normal values, desirable values, and values which are highly suggestive of brain at risk of further ischemic or metabolic injury, where we perhaps need to escalate our management in an individual way. And it's in this group where we might consider using treatments that we know have side effects but may be of benefit in this group at very high risk of deterioration, and that's another way where we might be able to identify patients who could respond to hypothermia. And although evidence from these comes from associations between the targets and outcome, at present that's a reasonable basis for treatment when we have no other evidence available. Although clearly it's not a causal relationship, and we need evidence from clinical trials that assess the benefit of these monitoring techniques and thresholds for defining a treatment in this cohort of patients, and we await the results of the boost-free and bonanza studies which are looking at the use of brain tissue oxygen monitoring. And we've already talked about cerebral optimal cerebral perfusion pressure and the cogitate study is an example of that. But even in the absence of of new evidence from clinical trials, we have existing consensus, international consensus on how we should manage patients who have both oxygen and intracranial pressure monitoring, and again, the civic guidelines provide us with algorithms for treating patients in this context. And they define the different types of raised intracranial pressure and oxygen derangement that you can see, defining the tiers of therapy and how these might be optimized when you have this information at your hand. And these provide some novel but but rational inferences. One of these is that where you have both low intracranial pressure but evidence of brain hypoxia, you may in fact look to decrease ventilation, to increase arterial PCO2, to increase oxygen delivery to the injured brain as long as the intracranial pressure is not raised. And in other circumstances where you see a reduction in glucose substrate within the brain tissue, you might look for different methods to improve glucose delivery even by increasing systemic glucose levels, and these are detailed in these charts. But one thing you don't see on here is much further additional comment about the benefit or otherwise of temperature control in the cohort of patients. Moderate hypothermia or mild hypothermia, sorry, between 35 to 36 °C is still seen as a treatment option for tier 3 therapy. But it doesn't talk about the management of patients with fever and whether targeted normothermia is beneficial as a neuroprotective strategy in patients with vulnerable brains at risk from further ischemic or metabolic injury. And here we can look to some more recent guidance provided by expert consensus, and this was led, uh, and this publication was led by Andrea Lavigna, supporters we heard from the European Society of Intensive Care Medicine and the Neuro Anesthesia Critical Care Society of the UK. And this group looked at best practice recommendations for the use of targeted temperature control. And in summary, um, what they concluded, or one of the major findings of the of this group was to recommend that controlled normal fermia, that is with a target core temperature between 36 to 37.5 °C, was recommended in these patients within tier one and tier two before considering mild hypothermia in tier three. So recognizing the importance of hypothermia as a potential or targeted normothermia, sorry, as a potential neuroprotective strategy in these patients. So if I just quickly summarize um some of the messages that I've covered here in my brief talk, is that best practice would suggest that all patients with traumatic brain injury should have temperature monitored continuously to ensure fever is promptly detected and treated. Because fever in this um context, the context of severe TBI with vulnerable brains at risk of further metabolic or ischemic injury. May benefit from the prevention of fever. In addition, controlled normothermia is strongly recommended as a therapeutic option to facilitate ICP and cerebral perfusion pressure management in this cohort of patients. Also, we need to look at what we provide for these patients to ensure that we are personalizing the care that we provide them and individualize what are potentially harmful treatments, even within tier 1 and tier 2 management to ensure that we account for things that we know will differ within the individual nature of the patients. Different CPP targets are suggested for patients at the extremes of age. We should also target an optimal CPP based on either management or sorry, measurement of cerebral autoregulation at the bedside or looking at indices of cerebral physiology. And where available, we should use downstream events as a marker of cerebral well-being. These include monitoring of the partial pressure of tissue oxygen within the brain and also using bedside microdialysis to look at the adequacy of substrate delivery, glucose to the brain and what impact that has on metabolism, looking at lactate or the lactate peruve ratio. And finally, we must remember that all of our IC therapies in this context are not risk-free, and we should titrate therapy intensity to need. Tier 3 therapies like mild or moderate hypothermia should be reserved for patients who fail treatment with the earlier tiers of therapy. And ultimately to assess the benefit of mild and moderate hypothermia, we will need a further relevant clinical trial. Thank you, that's the end of my talk and I'm welcome to take questions from er people attending the webinar. Thank you very much Jonathan. Thank you Andrea. I think we have some minutes for the discussion. Uh, I have some questions for you and I'm very curious about the response. The first one is uh um if we have some treatments which can reduce the time of hyperaxia and the fever is associated with outcome, why don't we find an association with outcome of this treatment? I, I, I think the, that, that, and maybe Andrea will correct me, but my understanding is the studies that we've looked at so far, um, haven't really addressed that question. Um, I, I, I, I, I think, um, when you're comparing um an intervention to standard care, it doesn't control for what the standard care is. I, I think it would be difficult to tease out. Weather The intervention has managed to control fever or or not because that's not the question that those those studies were were were set up to answer. Uh, yes, uh, uh, I, I, I think in, in observational studies it's quite clear that the burden of fever is associated with the, um, worse outcome. In the recent well conducted study, the Polar study that Professor Coles just talked about, the, the control group had very strict control normothermia, and so I, I think that answers the question with the and the question is, should we. Uh, favor prophylactic hypothermia over control normothermia as a patient comes in with traumatic brain injury, and, and the, the answer, which I, we accept as conclusive is no, and therefore we, we are not advocating for uh uh prophylactic hypothermia, but is there evidence that fever is harmful? Yes, absolutely. The fact is we, we do treat fever and um and, and, and therefore, We, we might not see that signal in trials, and the polar in particular, they were very good in the control group in maintaining control nothermia, which should be standard of treatment. Thank you. The next question is uh uh which is uh your clinical practice in the management of fever? So you said you don't use any prophylactic uh therapies, but uh do you start with uh pharmacological agents or feedback systems or. What's your practice? John, do you want to take that, or? I think I, I, I, I, I can, well, I can start, um, I, I think, uh, the management of fever again it's individualized to the patient. Um, it depends, um, it, it, is, is is a patient at risk who we're monitoring intracranial pressure and managing via an ICP protocol, in which case, um, I, I, in my hands, I think we would be targeting with my sort of bias originally would be, would be targeting, uh, nomophobia to start off with, um. But if they develop fever beyond that, yes, we'd use antipyretics such as paracetamol or non-steroidals, but then I, I, I suspect we'd already be using active, um, temperature management systems to not only continuously monitor the temperature to to control it within a normal normal thermic range. But we wouldn't be doing that in all patients, so in patients that clinically we either weren't monitoring intracranial pressure or we were several more days down the line and we were less concerned about the, the risk of causing further injury or or swelling or ischemia within the injured brain, we may be more relaxed and actually trying to normalize temperature and accepting um higher values of of of of of of temperature in that context. Andrea, uh, do you want to add anything, or no, I, I agree with with what Professor Colz just said. The, in the context of precision medicine and individualized care, you would take the effectiveness and side effects of various treatments into account and decide the most appropriate treatment for the patient depending on the severity of their injuries. One of the big discriminants would be a patients that's self-ventilating, uh compared to the patient that's uh mechanically ventilated in a self ventilating patients, even if we know that the effectiveness of uh uh pharmacological treatment is. Uh, very weak, we would probably start with paracetamol occasionally we would try diclofenac infusions, which are marginally better than paracetamol, but we wouldn't escalate towards automated the use of automated devices in that context. Necessarily because shivering is often problematic and patients do not tolerate it well. It, it's quite different if you have a patient where, where you have evidence of of fever occurring and they're mechanically ventilated and you had a CT scan that morning that shows the brain is very tight, they might be at risk of herniation, then we would have a very low threshold uh in uh instituting uh uh cooling with uh automated devices. Thank you, very clear. Uh, we have, uh, at least other 3 questions. I hope we have time. So the first one I think is about more, uh, ICP control because, uh, here, uh, it says the problem for us is where the center is the tertiary center is not interested or not yet interested in, uh, uh, in admitting the patient and we, how do we know when we are at the desperate stage of ICP. And intracranial hypertension, given that we don't routinely use ICP monitors. Good luck. Yeah, good luck, yeah, that's a, that's an interesting scenario, um. I, I, I. Yeah, I mean if they it it's difficult because it, it, this is, this is tied up in terms of what the significance of the injury is, and I, and I guess we, we do have to take advice from our neurosurgical center in terms of how best to best to manage these patients. But if, if it's the scenario I'm assuming that you're discussing is a patient that does have evidence of traumatic injury on the scan. At the, at the, at the stage of the discussion, there's no. Immediate need to transfer the patient unless they deteriorate clinically. Um, so in that context they're probably not sedated and they're more of the awake spontaneously breathing patient that Andrea described. So in that case you would still be looking to prevent fever were it to occur using simple measures, and you wouldn't be sedating them. And using other treatments to control physiology and help control temperature and if you were, you perhaps should have repeated the CT scan and discussed the patient again with your neurosurgical center to get a clearer idea of what further management they suggested. I, I, I, difficult to say much more than that. I don't know. Andrea, what do you think? I, I, I agree. If, if, if they, they have a scan that is concerning, they should be managed with ICP and everything else. I, I think the temperature becomes a bit of a detail. Yeah. That's what I was trying to say in a more polite way, I guess. Yeah, uh. Here there's another question, so presumably a negative trial has as many patients who may come to harm as benefits. How might temperature control be harmed? How, how might temperature control cause harm, I suppose it depends whether you mean temper control or, or, or hypothermia. Um, I mean temp temperature control to nor to, to, to normal fermia should, should not cause harm, but, um, temperature control for inducing hypothermia, uh, you know, we, we, we know it can cause side effects, it, um. As Andrea says, you need to be clear about why you're using it. You shouldn't really be using it if there's evidence of, um, or there's no evidence to suggest you should use it in patients with evidence of infection or sepsis, because it's a natural host defense mechanism. Clearly it can become excessive and result in harm to some degree, but it's, but that's that that that's not the scenario that we should be actively lowering temperature in. But even going beyond that, in patients who you are actively cooling. That does reduce the host's ability to fight off infection and so it can lead to increased secondary infections. Hypothermia, we also know can cause other systemic complications such as coagulopathy. Severe hypothermia below 34, 35 degrees can cause cardiac arrhythmias. And although it does lead typically to some degree of peripheral vasoconstriction and optimizing blood pressure, if you get to below those levels of temperature, it can indeed induce hypertension. So, and there are other systemic effects in terms of of how it can affect other organ systems which can lead to harm. Absolutely, and, and those are side effects that we expect when the treatment is working as expected, but there are also many situations where, uh, for example, if you have a failure of a temperature probe and you have a powerful automated device that's being misregulated, you can overcool patients accidentally with sometimes catastrophic effects. This is relatively rare, but it happens if you have, uh, if you're instituting a treatment, and these are quite, well, these are very effective treatments, but they have a burden on the patient. So often if you want to maintain controlled nothermia or therapeutic hypothermia, you would have to sedate and ventilate the patient to facilitate that. So ventilator associated pneumonia or all the other complications that you might have associated with that. are going to be factored in as a biological cost of your treatment. Some other devices based on intravascular cooling devices will induce. Uh, thrombotic or thromboembolic events associated with with having a cooling element or line infections related to that cooling element that you wouldn't otherwise have. So these are effective treatments, but they have a biological cost as well as financial cost. And so from many perspectives, the need to reserve these treatments in terms of indication. Intensity and duration to the right patient is key. Uh, thank you, Andrea. Uh, on top of this, um, do you think that local system of, uh, cooling like a helmet, uh, or, uh, I mean, feedback system, but, uh, mm, with, uh, with local cooling, uh, effect, I think there are also the those which go in in the north. Do you think this can be, uh, effective and reduce the side effect? That, that's the idea. I'm, uh, I, I need to declare I'm directly involved with the company working on this neuronal which is a, a, a caller, and we are starting a, a small trial, Celetherm to, uh, to provide selective uh uh uh brain cooling, and the idea would be, The main concern with systemic cooling is, is twofold. The first is the one that John described, which is essentially chest infection and, and worse ventilator associated pneumonia when you have systemic cooling. And the other one is that if you have a bulky system, a total body cooling system that's not transportable. You're often moving patients at the time, for a, for a, for example, to take them for a for a scan or to theater at the time where the brain is most vulnerable, and you typically disconnect them from this control, uh, temperature control devices having fluctuations at the time where the patient is most vulnerable. So the idea of having a a kind of smaller um selective device, be it a collar, a helmet, intranasal devices, that's portable. Uh, has two promises, one, to, sort of be transportable at this time when the brain is vulnerable, but also to keep a cool head and a warm heart, uh, so that you maximize neuroprotection whilst minimizing, uh, systemic complications. But this is still to be proven. OK, thank you. Uh, next question is which temperature should we take as a target for hypothermia, uh, both cerebral or systemic? Well, I, I, I, I guess you, you, you need, you need to measure both, um. If is the short answer. One, you, you may target your treatment based on brain temperature management, seeing that's the organ of interest here in terms of uh neuroprotective strategy, but, um, you probably need to correlate that with the effects on systemic, not, not just because, um, if you're using a targeted measure like Andrea has just mentioned, you, you want to know what the differential is, but, but equally you can get false readings and, uh, at least if you're getting a measurement of core temperature from two sites, you should get an idea. Of, of whether um you're hitting your desired target or whether um there is a measurement error um involved, um, so I, I, I, I think, although I would ideally um center my management on brain temperature, I think I would still continue to have an additional measurement of core temperature outside the brain. Andrea, anything to add or uh yeah, that's another topic that's very dear to me. The, there is surprisingly little high quality evidence on what are the variables effectively affecting the sort of brain to core, the brain to esophageal or brain to bladder temperature. And uh uh we are seeing uh uh we've been looking at it more in a more systematic fashion in the past couple of years. We see that there is a lot of variation there. So I think looking at both uh as John says, is key one from, from this patient safety perspective, it allows you to have a validation of reading so that you know that. You protect yourself from temperature monitoring failure, which is, as we said, particularly important, especially when using automated devices. But 2 is that you can gauge effectiveness of your selective. Brain cooling strategies and in fact, brain temperature and the burden of brain fever will be is one of the main endpoints that we are looking at when evaluating these selective temperature management devices. Thank you. Very, very, very last question, are there any conditions when you don't treat fever? I was thinking about a patient not in the acute phase where we have wind the sedation, does not respond to pharmacological therapies, so we have to choose between letting him uh with uh 38 degrees or using a feedback system and probably needing some sedation. What are you gonna do? I think I, I mean that's a, that's a very good question because that that's not an uncommon occurrence in terms of, you know, uh, probably, you know, more than a few days down the line, sort of 2nd week or more post, uh, management on the ICU with severe traumatic brain injury, you've had prolonged ICTP therapy, when you withdraw all of this, um, you quite often see p pyrexia unrelated to infection. Um, but in the scenario that that you're looking at there, if you're confident that there's no ongoing risk of further brain injury, cerebral edema has settled, um, CT scan doesn't suggest any worrying signs, you probably would move to a phase where you start to accept some of these derangements in a sort of gradual way and try and back off and reduce what you're doing, um, because it probably will not be a benefit at that time, but that's a clinical decision. Based on the imaging findings and the other monitoring markers, um, in terms of you've moved past this phase, when ongoing further injury to the brain is likely to occur. And we do do run such a kind of such clinical experiment when patients are discharged to the ward, and some of them develop a chest infection. Imagine after two weeks or a longer stay even on the ICU they are extubated and sent to the ward. One of the, I would say that the probably the most common cause for them being readmitted to the unit is that they would develop a chest infection and become very unded. Because even the healthy brain becomes kind of we all experience brain fog with a few lines of temperature. In these patients that have an acute brain injury, sometimes they do end up being reintubated and readmitted to the unit, but most of them don't, they tolerate it, so I, I think we do tolerate permissive yrexia in that phase. Thank you. I think uh we are running out of time, so it's time to wrap up. Uh, I would like to thank again uh SICM NBD for this webinar and especially I would like to thank, uh, both of the speakers, Andrea and Jonathan, and, uh, see you to the next, uh, event. Thank you. Thank you very much. Thank you, John. Thank you very much. Created by
