Good morning, everyone. This is Dr. Balala. I'm a pediatric intensivist at Driscoll Children's Hospital, Corpus Christi, Texas, USA. I'm also a clinical associate professor for Texas A&M University in College Station, Texas and University of Texas Medical Branch, Galveston in Texas. I'm also currently the research advisor for Driscoll Health System and also the founding chair of research task force at Driscoll Children's Hospital. I'm really honored and privileged to be invited by Society of Critical Care Medicine, especially the Current Concept Committee, for delivering the talk on neuroprotective strategies. Again, thank you so much for the Current Concept Committee for inviting me and let us go through the discussion on neuroprotective strategies. The topic of neuroprotective strategies is broad, especially in the Pediatric Intensive Care Unit, where we see many mechanisms of injury, including traumatic and non-traumatic injuries. This presentation covers common scenarios. We will begin by describing neuroinjury in children, then outline mechanisms of secondary neuronal injury. Neuroprotective strategies, both conventional and novel, will be identified, and we will review the evidence and approaches to targeted temperature management. How does neuroinjury occur? The initial injury to brain initiates a cascade of biochemical, cellular, and molecular events, and that cascade produces secondary neuronal injury. Biochemical, molecular, and cellular pathways all contribute to this secondary injury through mechanisms such as excitotoxicity, oxidative stress, endotheliopathy, and neuroinflammation. For example, after cardiac arrest, excitatory neurotransmitters are released from dead neurons, stimulating the surrounding viable neurons. That leads to a neuronal energy imbalance, energy failure, and subsequent neuronal apoptosis and necrosis. Oxidative stress occurs as a result of an imbalance between the production and elimination of free radicals, and that stress can lead to both immediate and delayed cell death. An increase in free radicals affects the membrane permeability, increasing leakiness through oxidation of the membrane lipids. Endothelial damage occurs as a result of both initial insult and ischemia reperfusion. The damaged endothelium triggers a coagulation cascade, resulting in microvascular thrombosis and tissue hypoperfusion, which in turn increases inflammation and neuroinjury. Endothelial damage also contributes to the disruption of blood-brain barrier causing cerebral edema. Finally, neuroinflammation, which is caused by microglia and astrocytes, leads to cytokine release, tissue damage and swelling, and increased intracranial pressure. What are the conventional neuroprotective strategies? In the intensive care unit, secondary neuronal injury can be prevented or minimized through tight control of temperature, oxygen, carbon dioxide, glucose, and sodium, along with seizure prevention and management. We will take these strategies individually. First, let's look at oxygen. Animal studies have shown that hyperoxia after return of spontaneous circulation contributes to oxidative stress, which may potentiate the post-resuscitation syndrome, while some adult studies have shown associations between hyperoxemia and increased mortality. Small observational studies of pediatric in-hospital cardiac arrest and out-of-hospital cardiac arrest have not shown any association between elevated partial arterial oxygen pressure and outcomes. A larger observational study of 1,427 pediatric survivors of in-hospital and out-of-hospital cardiac arrest showed that after adjustment of co-confounders, normoxemia after return of spontaneous circulation was associated with improved survival to pediatric intensive care unit discharge compared to hyperoxemia after circulation. Therefore, it may be reasonable for rescuers to target normoxemia after return of spontaneous circulation. Because an arterial oxyhemoglobin saturation of 100% may correspond to arterial partial oxygen pressure measuring between 80 and 500, it may be reasonable, when the necessary equipment is available, for rescuers to wean oxygen to target an oxyhemoglobin saturation of less than 100%, but greater than 94%. The goal is to achieve normoxemia while ensuring that hyperoxemia or hypoxemia are avoided. Ideally, oxygen will be titrated to a value appropriate to specific patient condition. What about ventilation? Adult data have suggested an association between hypocapnia and worse patient outcomes after return of spontaneous circulation following cardiac arrest. In other types of pediatric brain injury, hypocapnia is associated with worse clinical outcomes. One small observational study of both pediatric in-hospital and out-of-hospital cardiac arrest demonstrated no association between hypercapnia or hypocapnia and outcome. However, another observational study of pediatric in-hospital arrest revealed that hypercapnia was associated with worse survival to hospital discharge. It is recommended that arterial partial oxygen pressure be maintained tightly within the normal range. Severe hypercapnia and hypocapnia should be avoided. Information about carbon dioxide management after traumatic brain injury is plentiful. Use of hyperventilation was based on assumption that hyperamia was common after pediatric brain injury and that it reduced intracranial pressure by reducing luxury perfusion. However, more recent pediatric studies have shown that hyperventilation can produce hypoperfusion or cerebral ischemia. After traumatic brain injury, hypocarbia has also been shown experimentally to reduce the buffering capacity of the cerebrospinal fluid. What does that mean? When carbon dioxide is acutely reduced below 30, the buffering capacity is altered in a way that even a slight increase in arterial partial oxygen pressure can cause exponential increase in intracranial pressure. Therefore, ventilation targeting normal arterial levels of carbon dioxide, i.e., between 35 and 45 mmHg, is currently recommended. In patients with traumatic brain injury, prophylactic severe hyperventilation to an arterial partial carbon dioxide pressure below 30 in the initial 48 hours after injury is not suggested. If hyperventilation is used in the management of refractory intracranial hypertension, advanced neuromonitoring for evaluation of cerebral ischemia is appropriate. Myocardial dysfunction and vascular instability are common following resuscitation after cardiac arrest. Small observational studies involving pediatric in-hospital and out-of-hospital arrests demonstrated poor outcomes when children were exposed to hypotension during the post-arrest period. After return of spontaneous circulation, it is recommended that parenteral fluids, inotropes, or vasoactive drugs be used to maintain a systolic blood pressure greater than fifth percentile for age. When resources are available, continuous arterial pressure monitoring is recommended to identify and treat hypotension. Several studies have demonstrated an association between intracranial hypertension, systemic hypotension, and poor outcome after severe traumatic brain injury. Cerebral perfusion pressure is the most readily available correlate of global cerebral perfusion. You calculate this by subtracting the mean intracranial pressure from the mean arterial blood pressure. What do we know about glucose? The severity and duration of hyperglycemia following brain injury are consistently associated with worse outcomes. The data are insufficient to recommend for or against tight glucose control for children with severe brain injury and persistent hyperglycemia. The American Heart Association guidelines on hyperglycemia control following cardiac arrest in adults recommend maintaining a blood glucose level between 144 and 180 micrograms per deciliter. Next, sodium. It's important to avoid hyponatremia after brain injury. In a study, patients with traumatic brain injury who were maintained a sustained sodium level of 160 millimoles per liter or greater were less likely to have a favorable glass glucose outcome score of 4 or 5. Ideally, the sodium level should be maintained between 140 to 150 millimoles per liter after brain injury. Convulsive and non-convulsive status epilepticus occurs in a substantial proportion of critically ill children following brain injury. Status epilepticus increases the metabolic demand of neurons and causes neurometabolic dysfunction, which in turn contributes to secondary neuronal injury. This is one reason why continuous electroencephalography is so important in patients following brain injury. Although clear evidence to suggest that control of seizures improves outcomes in comatose survivors of brain injury, its management of status epilepticus after severe brain injury should follow American Academy of Neurology guidelines. This diagram depicts relationship of hypothermia and fever with the spectrum of neuronal injury after cardiac arrest. Hypothermia shifts cellular injury from necrosis to apoptosis to recovery, whereas fever tends to worsen the degree of neuronal injury from apoptosis to necrosis. Fever worsens the neuroinflammation, whereas hypothermia blunts neuroinflammation. Shankaran and colleagues conducted a study where they randomized 208 neonates with neonatal hypoxemic ischemic encephalopathy to receive either usual care or whole-body cooling. For cooling, they used a goal temperature of 33.5 degrees centigrade, which was maintained for 72 hours using a whole-body cooling blanket. They were able to demonstrate an 18% decrease in mortality and neurodisability in the infants who were in whole-body cooling group compared to usual care control group. In 2002, two randomized clinical trials showed improved neurological outcome in adults who were unconscious following an out-of-hospital cardiac arrest when they were cooled to between 32 and 34 degrees centigrade for 12-24 hours shortly after return of spontaneous circulation. The hypothermia after cardiac arrest trial or the HACA trial found a substantial decrease in mortality in patients treated with mild therapeutic hypothermia. These trials included only patients with initial ventricular fibrillation and ventricular tachycardia rhythm. Most adult institutions have adopted a targeted temperature management plan of 32-34 degrees usually for 24 hours. This was protocol used in these early trials. Recently, a French expert panel published guidelines on targeted temperature management for adult survivors of cardiac arrest. To develop the guidelines, the panel members evaluated eight meta-analyses, four systematic reviews, and several multi-center and single center trials and provided guidelines and expert opinions. Targeted temperature management is recommended to improve survival in adults who are resuscitated from out-of-hospital cardiac arrest with shockable cardiac rhythm and who remain comatose after return of spontaneous circulation. The guidelines also suggest that targeted temperature management should be considered in comatose adult survivors of out-of-hospital arrest with non-shockable cardiac rhythm. In pediatric patients, a retrospective single center study by Fink and colleagues showed no significant difference in mortality and neurologic outcome for children treated with either normothermia or therapeutic hypothermia after arrest. Now let's look at results of the therapeutic hypothermia after pediatric cardiac arrest trial. This was the largest randomized controlled pediatric trial for evaluating the efficacy and safety of hypothermia for neuroprotection after cardiac arrest. The trial enrolled more than 620 children over a six-year period at about 37 clinical centers throughout the US and Canada. These children ranged in age from 48 hours to 18 years and they were all comatose survivors of cardiac arrest who received at least two minutes of cardiopulmonary resuscitation. The patients were randomized to either received hypothermia or normothermia. The study used a surface cooling blanket to maintain the therapeutic temperature within the goal range. The enrollment, randomization, and initiation of targeted temperature management all were done within six hours of onset of cardiac arrest. The therapeutic hypothermia group followed the protocol outlined here. The normothermia group maintained the target range for the same duration, 120 hours, so a normal five-day protocol. The authors looked at both survival and functional outcomes at one year. What this trial showed was that compared with therapeutic normothermia, therapeutic hypothermia did not confer a significant survival benefit with good functional outcome at one year in comatose children who survived in-hospital or out-of-hospital cardiac arrest. What could be the possible cause behind these findings? One reason could be what happens during the rewarming phase. A resurgence of inflammation during rewarming may contribute to a loss of neuroprotective effects of therapeutic hypothermia, especially when rewarming occurs rapidly or in an uncontrolled manner. Three studies looked at the effects of rewarming following therapeutic hypothermia. In a swine model of neonatal hypoxic ischemic encephalopathy, investigators were able to demonstrate that rewarming induced cortical neuronal apoptosis. Another study found that rewarming after hypothermia following cardiac arrest shifts the inflammatory balance, resulting in release of inflammatory cytokines in the cerebrospinal fluid. Another swine study used the same rewarming protocol as the therapeutic hypothermia after pediatric cardiac arrest trial. When the microglia of the euthanized swine were examined, the investigators found that the microglia of the subjects in the neopneumothermia group had the appearance of resting microglia while the microglia in therapeutic hypothermia group exhibited larger cell bodies and shorter processes, meaning active inflamed microglia. These are the recommendations for post-cardiac arrest temperature management published by American Heart Association. For infants and children remaining comatose after cardiac arrest, it is reasonable either to maintain a five-day of continuous normothermia that is 36 to 37.5 degree centigrade or to maintain two days of initial continuous hypothermia that is 32 to 34 degree centigrade followed by three days of continuous normothermia. The liver should be aggressively treated after ROSC. The question is, should we cool or should we not cool? To some degree, the answer depends on patient's age and physiologic factors, as well as the protocol in place at your own institution. No matter what that protocol is, it's important to have one in place and to stay within that protocol or range, monitoring the patient and regulating very tightly during the five-day period. Overall, targeted temperature management with either therapeutic hypothermia or normothermia is a helpful strategy in children who remain comatose after return of spontaneous circulation. How do you decide whether to use hypothermia or normothermia? Therapeutic hypothermia could be considered in neonates with moderate to severe hypoxic ischemic encephalopathy and in absence of contraindications. That's because there is significant evidence from the neonatal literature that cooling is helpful in this population. That may be because the mechanisms in neonates are different from those seen most often in older children. We don't know. If we extrapolate the adult data, targeted temperature management of 32 to 34 degrees centigrade could also be appropriate for adolescents and young adults with ventricular fibrillation or ventricular tachycardia arrest and absence of contraindications to the cooling. What about the use of normothermia? In comatose children following asphyxial cardiac arrest, it should be considered. This is especially important in the absence of a protocol where the temperature may not be monitored as frequently and the patient may drift into febrile range. On the other hand, in comatose children following ventricular fibrillation or ventricular tachycardia cardiac arrest, one may consider using therapeutic hypothermia. The timing of initiation of a targeted temperature management strategy is also very important. Generally, the earlier it is initiated, the better it is. Even though early initiation is challenging, aim to begin within the first six hours after achieving spontaneous circulation. What about the duration of targeted temperature management? There is still no clear consensus on duration of cooling. The recommendation in adults is 24 to 48 hours of cooling. In neonates, a longer duration is sometimes considered. However, the protocol used in the therapeutic hypothermia after pediatric cardiac arrest trial specified a duration of 24 to 48 hours at a temperature between 32 to 34 degrees centigrade followed by slow re-warming of 0.5 degrees to reach normothermia and maintain normothermia for next three days or total five-day course. What are some of the practical considerations? First, multiple methods can be used to achieve and maintain temperature in a target range surface versus internal cooling, head versus whole body cooling and intravenous versus intranasal cooling. Intranasal cooling process is used only for adults and employs specially designed central lines for rapid delivery of the coolant. Intranasal cooling is also experimental at this time, although animal models show promise. The most common form of surface cooling is the cooling blanket. The blanket is placed below the patient and cold water circulates to and from the water reservoir through tubing within the blanket. When using these blankets, you must minimize the shivering response. That means the patient may need to be appropriately sedated and or paralyzed. It's important to closely monitor the patient's physiologic parameters and EEG. A second blanket may be placed on top of the patient to achieve more rapid cooling. The automatic mode is safer and easier for achieving and maintaining target temperature and it allows slow rewarming. The automatic mode also reduces nurse workload. A simpler approach is use of ice packs and cold water packs or rubbing or applying cold water to the body surface. The major drawbacks to this approach are lack of temperature control with a potential for over or under cooling. When using a cooling blanket, the team should continuously monitor the patient's core temperature using an esophageal or rectal thermometer probe or a bladder catheter in non-aneuric patients. Auxiliary or oral temperature measures are inadequate in these patients and tympanic temperature probes are rarely available and often unreliable. The probe is connected to the blanket in automatic mode. The blanket senses the patient's temperature and adjusts the temperature of the coolant as it circulates through the tubing inside the blanket. Let's outline a practical approach to TTM or Targeted Temperature Management. First, once a patient is identified as a candidate, begin induction as soon as possible, preferably within 6 hours of spontaneous circulation. Achieve a target temperature as soon as possible. For therapeutic hypothermia, maintain the core body temperature between 32-34 degree centigrade for 2 days. Follow this with slow re-warming at 0.5 degree centigrade every 4 hours until the patient is normothermic and then maintain normothermia for 3 days. So the total duration of the targeted temperature management protocol is 5 days. For a therapeutic normothermia strategy, maintain the core body temperature between 36 and 37.5 degree centigrade for 5 days beyond return of spontaneous circulation. Continuous measurement of temperature during this period is recommended and fever should be aggressively treated after return of circulation. Monitor the temperature, hemodynamic parameters, fluid electrolyte balance, renal function and lower temperatures are associated with hypokalemia and other electrolyte deficiency and therefore potential for kidney injury. Monitor the patient's end-organ function very strictly. Keep an eye out for infection, while lowering the temperature or avoiding fever, it may be more difficult to discern if an infection is present, so look at other parameters of infection. Provide optimal sedation and paralysis to avoid shivering. Monitor neurologic function, preferably with continuous EEG monitoring. Avoid enteral feedings and provide parenteral nutrition, hypothermia slows down the gut and therefore it can induce bowel hypoactivity and feeding intolerance. Follow the American Heart Association post-cardiac arrest care guidelines for prevention and management of multi-organ dysfunction. Finally, here is a quick summary of some of the newer neuroprotective strategies. Some of these have only been tested in rodents, including cannabinoids, anti-tumor necrosis factor alpha, minocycline, vitamin D, melatonin and erythropoietin. The others have been tested in adult patients and statins have been tested in both rodents and adults. Some of these agents have shown potential benefit, but there is no clinical evidence for their therapeutic use. Unfortunately, we haven't yet found the right molecule in these agents. However, moving forward, research may discover some of these molecules can be used as adjunct to targeted temperature management strategy.

neuroprotective strategies

secondary neuronal injury

therapeutic hypothermia

pediatric neuroinjury

seizure management

physiologic monitoring