Abstracts From the 2nd Annual Therapeutic Hypothermia and Temperature Management Conference March 1–2,2012 Miami,FL

Controversies in the Use of Hypothermia to Target TBI

David O. Okonkwo

Department of Neurological Surgery, University of Pittsburgh Medical Center, Pittsburgh, PA.

To date, 23 clinical trials of hypothermia treatment involving 1614 patients with severe traumatic brain injury (TBI) have been conducted. Results have been, overall, inconsistent. The two U.S. prospective randomized clinical trials of hypothermia in TBI (NABISH:1 and NABISH:2) failed to confirm the utility of hypothermia as a primary neuroprotectant for severe TBI grouped as a whole. In combining the data from both NABIS:H1 and NABIS:H2, retrospective subgroup analyses of all trial data revealed that hypothermic patients undergoing surgical evacuation of intracranial hematomas had better neurological outcomes than patients treated with normothermia. It makes far more intuitive sense that hypothermia would be beneficial to this specific subset of TBI patients with ischemia-reperfusion injury than to the global, heterogeneous population of severe TBI patients. Furthermore, new data from the Co-operative Study on Brain Injury Depolarizations group has brought forth two critical issues: 1) cortical spreading depression occurs in 55–60% of TBI patients requiring emergent craniotomy for evacuation of intracranial mass lesions and is associated with significantly worse patient outcomes; and 2) hypothermia can reduce the incidence of cortical spreading depression in the injured brain. Thus, one may speculate that a key pathophysiologic mechanism for hypothermia to target in TBI is cortical spreading depression. No Phase III clinical trial investigating any intervention in traumatic brain injury has ever been successful. In the past, TBI clinical trials have simply used the GCS score as the primary inclusion criterion and ignored the heterogeneity of TBI in trial design. To date, no trial to investigate specifically the effect of hypothermia on patients undergoing emergent surgical evacuation of intracranial hematomas has been performed. Nor has any study to target cortical spreading depression with an intervention in human TBI been attempted. Our critical field of TBI is primed for a prospective clinical trial of a well-established intervention (hypothermia) targeting a specific, known pathophysiology (cortical spreading depression).

Hypothermia: The Bedside Perspective

Mary Kay Bader

Mission Hospital, Mission Viejo, California.

The use of mild hypothermia and temperature control in the critical care setting are becoming an integral component of care. Manipulating a patient's core temperature impacts every body system and the team must understand the physiologic impact of lowering and raising temperature. From the “cold as ice” state to the return to normothermia, the critical care team implements a hypothermia protocol which includes management of every body organ as well as neuro-chemistries/electrolytes in order to reverse the anoxic hit which occurred during the cardiac arrest or neurologic disorder. Upon return to normal temperature, the critical care team must control and maintain normothermia. This is challenging to the team since shivering is a common side effect of temperature management. This presentation will focus on the development of evidence based protocols related to hypothermia and normothermia in order to reduce the potentially devastating consequences of injury. Translating research requires the team to work collaboratively and target interventions that impacts the system, team and patient population. Strategies will be highlighted on the key components for developing a protocol that the team can embrace and implement across multiple clinical arenas in the hospital. The application of hypothermia will be examined using protocols, challenging physiologic responses by patients and solutions to manage these responses. Using actual case studies, the participant will discover how unique patient populations respond to the implementation of therapeutic temperature management.

Intra-Operative Warming/Normothermia in Burn Patients

Carl I. Schulman

DeWitt Daughtry Family Department of Surgery, University of Miami Miller School of Medicine, Miami, FL

Maintaining burn patients' body temperature during surgery is a significant challenge. While increasing the ambient operating room temperature and other passive rewarming methods help, such measures have limited effectiveness and prove taxing on operating room personnel. We hypothesized that an intravascular warming catheter will help maintain patients' core body temperature during excision and grafting procedures. A retrospective case-control study in patients with major burns between January 2006 and June 2011. Cases receive an intravascular warming catheter, while controls receive traditional temperature conserving interventions. The intravascular warming catheter utilizes a closed loop of sterile saline circulated through a femoral central access. The saline bath temperature was automatically adjusted to maintain patient core temperature. As warming was maintained, the room temperature was gradually lowered to normal. Twenty-three patients were involved in 31 total cases using the catheter, compared with 39 controls in 62 surgeries. The mean temperature deviation for each catheter group was −0.76±1°C, and −0.80±0.9°C for the control group. Given 20-minute intervals throughout the operations, the mean patient temperature for cases and controls never deviated by more than 1°C. Operating room staff satisfaction has improved with decreased room temperatures. An intravenous warming catheter reliably maintained patient core body temperature during surgery. To date, this is the largest cohort study of such a catheter among burn patients. This system may be more effective than current warming techniques, with the potential to decrease the total number of procedures and decrease the time to complete wound closure.

Management of Complications in Therapeutic Hypothermia After Cardiac Arrest

Justin Lundbye

Hartford Hospital, Hospital of Central Connecticut, University of Connecticut, Hartford, CT.

Although evidence from both laboratory and patient studies have demonstrated a decrease in platelet function and a decrease in clot formation-time in patients undergoing therapeutic hypothermia (TH), this has not translated into a statistically significant risk of bleeding. In contrast, some studies have demonstrated mild increased risk of bleeding in patients undergoing PCI while receiving TH. Bleeding is generally not increased if there is no vascular compromise. Evidence shows that infection is substantially increased in patients who are admitted to the ICU after cardiac arrest with or without TH. Studies have reported that pro-inflammatory cytokines are affected by TH. Observational studies have also reported increased risk of infection in patients who receive TH after experiencing out of hospital cardiac arrest. The most common site of infection is pneumonia. Potassium disorder is common in patients undergoing TH and the cause is likely multifactorial. Up to 18% of patients undergoing TH may develop severe hypokalemia defined as less than 3.0 millimoles per liter. In this patient population, the risks of arrhythmia are increased. Pharmacokinetic changes during TH is common, however the clinical significance of this has not been well defined. Both animal models and other studies have demonstrated that ICU-type drugs such as propofol and fentanyl have increased serum concentration during the cooling phase. Cardiovascular disorders related to TH are common, although probably beneficial for the patient. Mild bradycardia can develop, though seldom requires intervention. Cardiac output is decreased in patients undergoing TH, but the net effect after reducing the oxygen demand by the tissue translates into a positive outcome.

Long-Term Benefits of Early Cooling in the Post-Cardiac Arrest Patient

Hans Friberg,^1,2 Tobias Cronberg,^2,3 and Niklas Nielsen^2,4

¹Department of Intensive- and Perioperative Care, Skåne University Hospital, Lund, Sweden.

²Department of Clinical Sciences, Lund University, Sweden.

³Department of Neurology, Skåne University Hospital.

⁴Department of Anesthesiology and Intensive Care, Helsingborg Hospital.

There is strong evidence from animal experiments that hypothermia is neuroprotective and that early cooling after an ischemic insult is superior to a later onset. It has also been shown that recovery after an ischemic brain injury occurs during a prolonged time, and assessment of outcome should therefore be postponed. Induced hypothermia after cardiac arrest has been incorporated in clinical guidelines and implemented in post-resuscitation care. However, its benefits are still questioned and the optimal cerebral outcome measure after cardiac arrest is not known. There are several publications, but little information that expands our knowledge on important outcomes after cardiac arrest, such as cognitive function and quality of life. Moreover, the optimal time to evaluate neurological function is not known. The ongoing Target Temperature Management after cardiac arrest trial (TTM-trial) (Clinical Trials, in press) will hopefully shed some light on these issues. There is a consensus that cooling should be initiated as early as possible after cardiac arrest, but there is no human data supporting this recommendation. In a large, retrospective study, no association was found between timing of cooling and outcome. Other studies have shown that a lower body temperature on admission and a faster decline in body temperature to below 34°C were associated with a worse outcome, suggesting impaired thermoregulation in the more severly injured patients. Finally, while animal data suggest that intra-arrest cooling improves the rate of ROSC and survival, this still needs to be shown in humans.

Regionalized Post-Resuscitation Care for Out-of-Hospital Cardiac Arrest

Joseph P. Ornato^1,2

¹Department of Emergency Medicine, Virginia Commonwealth University, Richmond, VA.

²Richmond Ambulance Authority, Richmond, Virginia.

Out-of-hospital cardiac arrest claims the lives of approximately 382,800 adult Americans each year. The American Heart Association (AHA) has recently issued a Policy Statement⁴¹ calling for the establishment of regional systems of post-resuscitation care to manage cardiac arrest patients with a “care bundle” that includes therapeutic hypothermia, cardiac catheterization/percutaneous coronary angiography for ST-elevation myocardial infarction, electroencephalographic monitoring (preferably with continuous EEGs), and goal-directed management of hemodynamics, oxygenation/ventilation, glucose, and metabolics (i.e., electrolytes, renal function). Virginia Commonwealth University (VCU) in Central Virginia has developed such a regionalized system involving 10 community hospitals functioning as hypothermia “induction centers” as the leading edge of VCU's Advanced Resuscitation Cooling Therapeutics and Intensive Care (ARCTIC) program. This effort is multidisciplinary and involves commitment by all providers in the continuum of cardiac arrest care from EMS personnel serving rural communities, to the nurses and physicians in the community hospitals and at VCU. An added value of the partnership has been the development of an education program in post resuscitation care provided at no cost to the rural healthcare providers by VCU in how to optimize post resuscitation care and outcomes. From 2008–10, approximately 30% of the 168 patients enrolled in ARCTIC were transfers from community hospitals. Survival to hospital discharge for patients who qualify for ARCTIC treatment averages 52%.

Efficacy and Mechanisms of Protection Against Hemorrhagic Stroke in Rodents

Frederick Colbourne, Shannon Wowk, Sarah Jones, and Yonglie Ma

Department of Psychology & Centre for Neuroscience, University of Alberta, Edmonton, Alberta, Canada.

Mild hypothermia is remarkably effective against global and focal ischemia. However, cooling is considerably less effective against intracerebral hemorrhage (ICH), at least in rat models of striatal hemorrhage created by infusing collagenase or whole blood. The generally weak and inconsistent histological and functional benefit contrasts with findings that hypothermia lessens edema, blood brain barrier dysfunction and inflammation in these ICH models. We tested whether systemic and brain-selective cooling treatments (e.g., one or three days at 33°C vs. a normothermia control group) lessen edema after intra-striatal infusions of thrombin (2.5 U) or iron (FeCl₂; 1 mM). Both iron and thrombin caused moderate increases in brain water content that was not attenuated by cooling. We also evaluated whether cooling affects the amount of non-heme iron in the brain after collagenase-induced ICH. As expected, non-heme iron levels rose considerably in the damaged hemisphere after ICH at 7 and 28 day survival times. Cooling did not have any appreciable effect on iron levels. In summary, these findings suggest that cooling does not protect against two key factors in secondary degeneration after ICH, namely iron and thrombin toxicity. Therefore, these experiments help to explain the weak neuroprotective effect against ICH, which contrasts with findings in ischemia models.

Update on the ICTuS 2/3 Trial: Therapeutic Hypothermia for Acute Stroke

Patrick D. Lyden, on behalf of the ICTuS 2/3 Trial Investigators

Department of Neurology, Cedars Sinai Medical Center, Los Angeles, CA.

The ICTuS (Intravascular Cooling Trial of Ischemic Stroke) began in 2002 with a pilot feasibility and safety trial. Subsequently, the ICTuS-L trial established feasibility and safety when combining endovascular cooling with thrombolytic therapy. In ICTuS-L however more patients suffered pneumonia in the treated arm. Also, cooling to target was delayed, creating the need for hypothermia induction with large volumes of intravenous cold saline during catheter placement. Therefore, prior to beginning a large, pivotal Phase 3 trial of endovascular hypothermia, a smaller Phase 2 trial is underway to assess 4 points of interest: pneumonia incidence; volume overload; interim futility and efficacy; and adequate recruitment rates. After 50 run-in patients, if FDA approves, ICTuS 2 will proceed to enroll another 400 patients. Patients all receive rt-PA intravenously. After randomization, hypothermia patients received 2L cold saline, 30 mg of buspirone, 1 mg/kg intravenous meperidine, and an endovascular catheter (Accutrol, Phillips/Innercool). Shivering is controlled with meperidine titration but if target cannot be reached without excessive sedation, the target is reset to a higher level (permissive hypothermia). Patients are examined daily for signs of pneumonia and aggressive work up, including prophylactic antibiotics, are required. Normothermia patients are monitored identically for pneumonia; fever is controlled with standard measures. Currently there are 11 active study sites. To date, 36 patients are enrolled, achieving the target rate of 0.4 patients/site/month. The DSMB approved the study to proceed after review of the first 29 patients and another DSMB review is planned. (Supported by NS P50xxxxx and P50YYYYY).

Current Study Results in Therapeutic Hypothermia in Acute Myocardial Infarction

David Erlinge

Department of Cardiology, Lund University, Skane University Hospital, Lund, Sweden.

Hypothermia is an established treatment after cardiac arrest to protect against cerebral injury. Here it is discussed if hypothermia can also be used to protect the heart during ischemia induced by ST-elevation myocardial infarction (STEMI). Mild hypothermia (32–35°C) may be of benefit as adjunctive treatment for STEMI by having positive effects during ischemia to reduce infarct size and on the four components of ischemia reperfusion injury: myocardial stunning, microvascular obstruction, reperfusion arrhythmias and lethal reperfusion injury. In the treatment to reduce cerebral injury after cardiac arrest hypothermia can be initiated after reperfusion and should be maintained for 24–48 h. In contrast, for heart protection evidence suggests that hypothermia should be initiated as early as possible during ischemia, at least before reperfusion. Clinical and experimental results indicate that reaching a temperature of less than 35°C before reperfusion is of paramount importance in order to reduce infarct size in the treatment of STEMI patients. Treatment after reperfusion can be relatively short. Hypothermia has wide-ranging effects on most of the mechanisms involved in ischemia and reperfusion injury which may explain the potent, highly reproducible cardioprotective effects seen in a large number of studies in different species. Cooling awake patients with STEMI is safe, feasible and well tolerated, but anti-shivering strategies must be used. Curently the CHILL-MI trial is recruiting 120 patients in a randomized clinical multicenter design and evaluates infarct size with MRI. Other smaller trials are also ongoing.

Ultra-Rapid Cooling Strategies for Treating Cardiac Arrest and Acute Myocardial Infarction

Kees H. Polderman

Pittsburgh University Medical Center, Pittsburgh, Pennsylvania.

Induction of mild hypothermia (32–34°C) after cardiac arrest improves neurological outcome. Earlier and more rapid induction of hypothermia may further improve outcome. Many side effects of hypothermia occur in the induction phase, and more rapid cooling may decrease the risk of such side effects. Finally, there is evidence that hypothermia applied before coronary intervention may reduce infarct size. Use in humans would require ultra-rapid induction of hypothermia, before coronary intervention. Currently there are four ways in which this could be accomplished: extracorporeal circulation, immersion in ice-cold water, high-capacity endovascular cooling combined with other methods (cold fluid infusion, ice-packs), and flushing of the peritoneal cavity with refrigerated fluids. The first two methods have been successfully tested in CA patients, with cooling speeds of 4–6°C/hr (extracorporeal circulation) and 3°C/hr (cold fluid immersion). These methods are likely to be too invasive (extracorporeal circulation) or uncomfortable (cold fluid immersion) for use in awake patients. Endovascular cooling combined with cold fluid infusion is currently being studied and early data suggest that cooling rates of 3–5°C could be achieved, with good maintenance and controlled rewarming in CA and AMI patients. We recently tested automated peritoneal lavage in 45 CA patients and 3 patients with AMI and achieved cooling rates of 12–13°C/hr, with successful maintenance and rewarming in both CA and AMI patients. In conclusion, endovascular cooling combined with additional methods and automated peritoneal lavage are the most promising methods to induce and maintain hypothermia in awake, non-intubated patients such as those with AMI or stroke.

Hypothermia Applications in Pediatric Neurocritical Care

Patrick M. Kochanek

Departments of Critical Care Medicine, Anesthesiology, Clinical and Translational Science, and Pediatrics, University of Pittsburgh School of Medicine, and the Safar Center for Resuscitation Research, Pittsburgh, PA.

The use of hypothermia in cerebral resuscitation has its roots in pediatrics related to the classical reports of remarkable outcomes in children who were victims of cold water drowning. Hypothermia then gained acceptance for application after the insult in the treatment of pediatric drowning and other pediatric conditions in the 1970–1980s. However, it subsequently fell out of favor. In this presentation, current work by investigators at the Safar Center for Resuscitation Research and the Children's Hospital of Pittsburgh focused on pediatric application of mild hypothermia in severe traumatic brain injury, cardiac arrest, and other uses in pediatric neurocritical care will be presented. The impact of hypothermia on mechanism of secondary brain injury in infants and children, the effect of hypothermia on drug metabolism in children, and discussion of current applications of mild hypothermia in pediatric cardiac arrest will be presented. Finally, thoughts about potential future applications of hypothermia in pediatric neurocritical care will be discussed.

Early Induction of Hypothermia for Evacuated Intracranial Hematomas: A Post Hoc Analysis of Two Clinical Trials

Guy L. Clifton, Christopher S. Coffey, Sierra Fourwinds, David Zygun, David O. Okonkwo, Alex Valadka, Kenneth R. Smith, Melisa L. Frisby, Richard D. Bucholz, Elisabeth A. Wilde, and Harvey S. Levin

University of Texas Medical School, Houston, Texas.

NABIS:H I was a multicenter clinical trial of 392 patients with severe brain injury randomized to normothermia or hypothermia for 48 hours. NABIS:H II was a similarly designed multicenter clinical trial of 97 patients but reaching 33°C in half the time as NABIS:H I. In NABIS:H II outcome was poor in 5 of 15 patients with evacuated hematomas in the hypothermia group and 9 of 13 patients with evacuated hematomas in the normothermia group (p=0.02). In NABIS:H II the latest that any patient randomized to hypothermia reached 35°C was 1.5 hours after surgery start. Applying these criteria to NABIS:H I, 31 of 54 hypothermia-treated patients with evacuated hematomas reached a temperature of 35°C or less within 1.5 hours after surgery start time, and the remaining 23 patients reached 35°C later. Outcome was poor in 14 of 31 patients (45%) reaching 35°C within 1.5 hours of surgery, 14 of 23 patients (61%) reaching 35°C more than 1.5 hours of surgery and 35 of 58 patients (60%) in the normothermia group (p=0.16). No confounding predictors of outcome were found. A meta-analysis of 46 patients with hematomas in both trials who reached 35°C within 1.5 hours of surgery start showed a significantly reduced rate of poor outcomes (41%) compared with 94 patients treated with hypothermia who failed to reach 35°C within that time and patients treated at normothermia (62%, p=0.009). Induction of hypothermia to 35°C before or soon after surgery start with maintenance at 33°C for 48 hours thereafter may improve outcome of patients with hematomas and severe brain injury.

The Utility of Hypothermia in Treating the Axonal and Vascular Damage Evoked by Experimental Traumatic Brain Injury (TBI)

John T. Povlishock and E.P. Wei

Medical College of Virginia Campus of Virginia Commonwealth University, Richmond, Virginia.

Although clinical studies have provided a relatively checkered assessment of the benefits of posttraumatic hypothermic intervention, laboratory studies continue to confirm its utility in attenuating the damaging sequelae of TBI. Here we revisit data that supports the utility of posttraumatic hypothermia while addressing new data that speaks to the benefits provided by combinational drug approaches incorporating hypothermia. In TBI-injured rats, we have consistently demonstrated traumatically-induced axonal change reflected in local axonal swelling as well as concomitant microvascular dysfunction. Both pathologies, which are also found in humans, respond positively to early posttraumatic hypothermia, with 33°C hypothermia achieving significant attenuation of the burden of axonal damage and the accompanying pial vascular dysfunction. This hypothermic protection was observed when the posttraumatic hypothermic intervention was followed by slow rewarming, with the caveat that rapid rewarming not only reversed any benefit but also, exacerbated the existing axonal and vascular pathology. In these studies, the time of initiation of the hypothermic intervention as well as its overall duration were important variables in influencing anysubsequent neuronal and vascular protection. Importantly, the use of post TBI hypothermia extended the therapeutic window of efficacy for known neuroprotective/vascular protective strategies. Hypothermia coupled to delayed therapeutic intervention, typically resulted in an improved axonal and vascular protection over that achieved via the use of either treatment strategy alone. Mechanistically, it appears that the combination of hypothermia with targeted drug therapy most likely slows the progression of the ensuing pathology, allowing for its better therapeutic targeting. Further, hypothermia appears to reduce drug metabolism and/or clearance, most likely providing further protective value.

A Prospective Intervention Study to Compare Efficiency of Different Cooling Technologies for Mild Therapeutic Hypothermia

Sonder P.,¹ Janssens G.N.,¹ Henry C.L.,¹ Beishuizen A.,² Rittenberger J.,¹ Callaway C.,¹ Girbes A.R.,² and Polderman K.H.¹

¹Department of Critical Care Medicine, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania.

²Department of Critical Care Medicine, VU University Medical Center, Amsterdam, The Netherlands.

Mild therapeutic hypothermia (32–34°C) is neuroprotective in patients with post-hypoxic brain injury after cardiac arrest. Various mechanical devices are available to induce and maintain hypothermia followed by slow and controlled re-warming. These devices have not been well studied in comparative clinical trials. We performed a prospective intervention study to compare four frequently used cooling systems. Multi-centered prospective intervention study in four university hospitals. Four different cooling systems were used to induce and maintain hypothermia. Two devices (Medi-Therm^® and Blanketrol^®) used external water-circulating cooling blankets, one (Arctic Sun^®) used gel-coated adhesive cooling pads, and one (ThermoGard^®) used endovascular cooling catheters with balloons circulating ice-cold saline. For the latter system we studied three different catheters with two, three or four water-circulating balloons, respectively. So far 70 patients have been enrolled. Baseline characteristics were similar for all groups. Endovascular cooling appeared to have the fastest induction rates but this was significant only vs. the Medi-Therm system (p=0.019). Time within target range±0.5°C was 96.8±7.0% for ThermoGard, 57.4±29.3% for Medi-Therm, 64.5±20.1% for Blanketrol, and 81.8±25.2% for Arctic Sun. The differences between ThermoGard vs. Medi-Therm, ThermoGard vs. Blanketrol, and Arctic Sun vs. Medi-Therm were statistically significant. No major complications occurred with any device. This study is ongoing and the results are preliminary. Based on the preliminary results endovascular cooling may provide faster induction, and cooling catheters and adhesive gel pads may provide more reliable temperature control during maintenance compared to traditional surface cooling devices.

A Prospective Intervention Study to Compare Efficiency and Reliability of Different Cooling Technologies for Fever Control

Janssens G.N.,¹ Sonder P.,¹ Henry C.,¹ Beishuizen A.,² Rittenberger J.,¹ Callaway C.,¹ Girbes A.R.,² and Polderman K.H.¹

¹Department of Critical Care Medicine, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania.

²Department of Critical Care Medicine, VU University Medical Center, Amsterdam, The Netherlands.

Symptomatic fever control is becoming an increasingly accepted goal of therapy in patients with neurocritical illness. Various mechanical cooling devices are available to achieve control of a patients' core temperature. Few comparative data are available on the efficacy of these cooling devices. Previous studies have not always used cooling devices at optimal capacity. To compare the efficacy and safety of three temperature management systems. Multi-centered prospective intervention study. Three university teaching hospitals (tertiary referral center). Mechanical cooling for fever control was induced by one of three cooling systems: external water circulating cooling blankets (Medi-Therm^®), water-circulating gel-coated adhesive pads (Arctic Sun^®) and intravascular cooling catheters (Thermogard^®). A total of 101 patients with various types of brain injury were included. A total of 118 cooling interventions were performed and analyzed for time to target temperature and fluctuations during the maintenance phase. We found that endovascular cooling had the fastest induction rate at 0.587°C/h. The time within target range was 87.2% for the Arctic Sun system, 71.2% for the ThermoGard system, and 73.0% for Medi-Therm blankets. There were significantly more sedated patients (81.0%±67.5%) in the ThermoGard group. There were differences within the endovascular group between different types of cooling catheters, with the lowest time within range (67.3%) for the Cool Line catheter and the highest time within range (98.3%) for the Quattro catheter. No major complications associated with cooling occurred in one of our patients. The new generation of cooling devices provide fairly accurate temperature control for fever when used as indicated. In our study endovascular cooling had the fastest induction rate while gel-coated cooling pads had the highest proportion of time within target range.