Abstracts from the 6 th International Hypothermia and Temperature Management Symposium September 12–14,2016 Philadelphia,Pennsylvania Hosted by Thomas Jefferson University
Available accessAbstractFirst published online August, 2016
Abstracts from the 6 th International Hypothermia and Temperature Management Symposium September 12–14,2016 Philadelphia,Pennsylvania Hosted by Thomas Jefferson University
Lawson Health Research Institute, London, ON, Canada
Robarts Research Institute, London, ON, Canada
Department of Medical Imaging and Biophysics, Western University, London, ON, Canada
Background: We developed a non-invasive, portable and inexpensive system for cooling the brain selectively immediately following cardiac arrest. The method was based on spraying cold air, generated by a vortex tube, into the nostrils at different flow rates using a feedback automatic controller. In this study, we evaluate the ability of this approach to tightly control the rewarming rate to be not more than 0.25°C/h in a pig model.
Methods: Experiments conducted on four pigs (2–3 months old, 28 ± 2 kg). Body temperature measured using an esophageal and a rectal temperature probe. Brain temperature also measured continuously and invasively with a thermocouple probe. Intranasal brain cooling initiated by setting the controller to the desired air temperature of −5°C and flow rate of 40 L/min using medical air. Once the brain temperature stabilized at the target temperature of 33 ± 1°C, the flow rate and air temperature adjusted automatically to maintain the target temperature while core body temperature maintained above 37°C. After 4 hours of cooling, the brain temperature allowed to gradually return to the baseline temperature.
Results: Following baseline, brain temperature decreased rapidly from 37.9 ± 0.3°C to 35.0 ± 0.8°C within 30 minutes and then stabilized at 33.5 ± 0.5°C within the one hour of brain cooling. Following four hours of cooling, the brain then allowed to gradually rewarm from 33.2 ± 0.2°C to the temperature of 35.1 ± 0.5°C in 6 hours corresponding to rewarming rates of 0.3 ± 0.1°C/h.
Conclusions: The method is able to tightly control the rewarming rate within 0.3 ± 0.1°C/h.
Shorter Cooling Time Is More Effective for Acute Ischemic Stroke
LambJessica1RajputPadmesh1LydenPatrick1
Cedars-Sinai Medical Center
Background: Therapeutic hypothermia (TH) shows considerable promise, but translation to clinical use has been problematic. Recent data suggests shorter, deeper cooling may be preferable over longer cooling times. Also, in contrast to prior rodent models using surface cooling, clinical TH often uses endovascular cooling technology. We modeled recanalization and simulated endovascular cooling in a rodent model.
Methods: A perivascular catheter was implanted retroperitoneally in 28 male, 300 g, Sprague Dawley rats 6 days prior to a 4 h heat blunted nylon filament MCAo. At reperfusion animals were randomized by an investigator outside the laboratory to brain temperature 33°C or 37°C (n = 7 to 8 per group) monitored by a temporalis muscle needle thermistor for 2 or 4 hours. Saline pumped through the closed loop cooling circuit convectively changed vena cava and body core temperature. After TH the circuit was disconnected and the animals returned to cages. Bederson neuroscores were obtained 4 h and 24 h after occlusion. After 24 hours, rats were sacrificed for TTC exclusion.
Results: Perivascular cooling achieved very rapid target temperature (<15 min) and maintained temperature within 0.09°C. Corrected infarct volume (TTC exclusion) mean ± SE were smallest after 2 hours 33°C TH: 14.4 ± 2.4 (n = 8) and largest after 4 hours 37°C 26.6 ± 2.6 (n = 5) (p < 0.01, ANOVA, Dunnett's). Volumes were intermediate after 4 hours 33°C 21.2 ± 2.6 (n = 7) or 2 hours 37°C 18.8 ± 2.3 (n = 8). Clinical scores were concordant.
Conclusions: The combination of recanalization after 4 h MCAo and only 2 h TH yielded highly significant neuroprotection, with a 50% treatment effect, compared to 4 h hypothermia or normothermia.
Background: Although hypothermia between 30°C and 36°C is powerfully neuroprotective during ischemia, clinical translation has proven elusive. Recent clinical data suggest prolonged cooling may be ineffective, but the mechanism is unknown. Astrocyte conditioned media (aCM) powerfully protects neurons from oxygen glucose deprivation (OGD). We hypothesized that prolonged cooling may inhibit protective astrocyte responses to ischemia.
Methods: We studied primary neurons and astrocytes cultured from E19. After 7 days, astrocytes were exposed to OGD ranging from 30 to 240 min at 33, 35 or 37°C after which conditioned media was removed and stored. Subsequently, neurons were treated with one of the OGD aCM and then subjected to normothermic 120 min OGD followed by normothermic reperfusion. After 24 h, cell viability was measured using 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT). Optical densities were normalized to cultures not subjected to OGD.
Results: Normothermic 120 min OGD killed most neurons (20% survival). Neuron viability was significantly improved (95% survival, p < 0.001) by aCM from normothermic 120 min OGD conditioned astrocytes. Target-depth 33°C aCM significantly ameliorated the protection (25% survival, p < 0.001). Target-depth 35°C aCM was intermediate (60% survival, p < 0.001). ACM after exposure to only 30 min OGD at 33°C gave 40% protective effect, compared to longer 33°C OGD exposed aCM (p < 0.001).
Conclusions: OGD aCM powerfully protects neurons during subsequent OGD, but this effect is eliminated by hypothermic treatment of the astrocytes in a graded, temperature dependent manner. These data may provide insight into the failure of prolonged cooling in clinical trials.
Ischemia Reperfusion Injury as a Modifiable Therapeutic Target for Cardioprotection or Neuroprotection in Patients Undergoing Induced Hypothermia
NicholGraham1MadathilRonsonMD1HiraRaviMD1ElrodJoAnn BroeckelPhD1StoecklMathiasMD2SterzFritzMD2
University of Washington
Medical University of Vienna
Background: Quick restoration of blood flow is essential in patients who have cardiac arrest or other conditions associated with local or global cessation of blood flow. This restoration of flow is associated with multiple deleterious cellular changes.
Methods: We review these changes to try to understand whether ischemia-reperfusion injury (RI) is a potentially modifiable therapeutic target for cardioprotection or neuroprotection in patients resuscitated from cardiac arrest.
Results: Induction of hypothermia (IH) involves cooling an ischemic organ or body. This has pluripotent effects that reduce the potential harm associated with RI in the heart and brain by reducing opening of the mitochondrial permeability transition pore. In contrast to prior trials, a recent trial of IH did not demonstrate these treatments to be effective. Lack of survival benefit was associated with a longer apparent time to target temperature between trials (about 11 hours vs. about 4 to 8 hours.) In a substudy reported by the largest enrolling site in this trial, the 33°C group received significantly more propofol than the 36°C while in their respective target temperature range (421 ± 177 versus 339 ± 175 mg/h; P = 0.005). Propofol has dose-dependent effects in mitochondria: At low doses (<100 microM), it reduces harmful reactive oxygen species. However at high doses (≥200 microM), it reduces ATP synthesis.
Conclusions: Concurrent interventions including time and propofol may attenuate the effect of IH on patients resuscitated from cardiac arrest.
Abstracts
The Impact of Cooling Techniques on Out-of-Hospital Cardiac Arrest; Surface Cooling Versus Intravascular Cooling
NakashimaTakahiro1TaharaYoshio1NoguchiTeruo1YasudaSatoshi1ArimotoHideki2NagaoKen3KurodaYasuhiro4HaseMamoru5ShiraiShinichi6NonogiHiroshi7
National Cerebral and Cardiovascular Center
Osaka City General Hospital
Surugadai Nihon University Hospital
Kagawa University Hospital
Sapporo City University Hospital
Kokura Memorial Hospital
Shizuoka General Hospital
Background: Target temperature management (TTM) between 32°C and 36°C for out-of-hospital patients (OHCAs) was established. However, there is no evidence which cooling technique is more effective for OHCAs.
Methods: This multicenter prospective cohort was conducted in 14 hospitals between 2005 and 2011. A total of 477 OHCAs underwent TTM, target temperature was 34 ± 1°C and maintenance time was 24–48 hours. Among 328 OHCAs excluded patients with extracorporeal cardiopulmonary resuscitation, we compared 192 patients who underwent surface cooling (Group-S) with 136 patients who underwent intravascular cooling (Group-I). We compared the quality of TTM and favorable outcome, defined as Cerebral Performance Category 1–2 at 3 months after collapse between the 2 groups.
Results: There were no significant differences of age, sex, witnessed, bystander cardiopulmonary resuscitation (CPR) and shockable rhythm between the 2 groups. Regarding the quality of TTM, in Group-I, time from cooling initiation to achieving target temperature was shorter (301 ± 261 minutes vs. 248 ± 195 minutes, p = 0.047) and the accuracy of management is significantly higher compared Group-S (93% vs. 53%, p < 0.001). After adjusting for age, sex, witnessed, bystander CPR, time from collapse to ROSC and initial rhythm of collapse, there were no significant difference in the rate of survival and favorable neurological outcomes between the groups.
Conclusion: The cooling techniques might not impact survival rate and favorable outcome in OHCAs. However, the system of intravascular cooling could shorten the time from cooling initiation to achieving target temperature and permit more accurate management.
Prolonged Hypothermia for Acute Ischemic Stroke and Increased Pneumonia Risk
LydenPatrick1RamanRema2ErnstromKarin2HemmenThomas3RappKaren3ParkerStephanie4GrottaJames4
Cedars-Sinai Medical Center
University of Southern California
UCSD
Memorial Medical System
Background: We developed an antishivering regimen during the ICTuS-L trial, but encountered a significantly elevated incidence of pneumonia. We sought to reduce pneumonia risk by implementing a precise definition of pneumonia, rigorous surveillance, nasogastric drainage, and restrained use of antishivering medications, notably meperidine.
Methods: The ICTuS-2/3 protocol has been published and sought to determine whether thrombolysis and hypothermia/antishivering regimen is superior to thrombolysis alone for acute ischemic stroke. All patients underwent daily surveillance using the CDC pneumonia definition, with prompt initiation of antibiotic therapy in suspected cases.
Results: 120 patients were enrolled (63 hypothermia (HY) and 57 normothermia (NT)). HY plus antishivering treatment had little effect on blood pressure or heart rate, but respiratory rate was significantly depressed, indicating over-treatment with antishivering medications. The incidence of serious adverse events was 41% in the HY group and 35% NT, OR (95% CL) 1.30 (0.58, 2.92). Mortality was 15.9% in the HY group vs. 8.8% in the NT, for an OR (95% CL) of 1.95 (0.56, 7.79). Pneumonia was confirmed in 19% in the HY group vs. 10.5% in the NT group, for an OR (95% CL) of 1.99 (0.63, 6.98). No serious adverse event trends in any other specific organ class appeared in the ICTuS-2 data.
Conclusions: There was a trend toward higher mortality in the HY group, perhaps related to a trend towards more pneumonia. Having eliminated case ascertainment bias, the elevated pneumonia rate in ICTuS-2 probably indicates a real side effect of prolonged (24 hours) HY plus antishivering medications.
A Biological Effect or an Imbalance of Co-Interventions? Exploratory Analyses of the Eurotherm Trial
PudduIreneMD1TacconeFabio S.MD, PhD2van der JagtMathieuMD, PhD3RodriguezAryellyMSc4BattisonClaireRGN4AndrewsPeter, J.D.MD, MB, ChB4
University of Turin
Hôpital Erasme Bruxelles
Erasmus MC Rotterdam
University of Edinburgh
Background: Traumatic brain injury (TBI) is a major cause of death and severe disability worldwide. The Eurotherm3235Trial was performed from November-2009 to October-2014, to evaluate the efficacy of titrated therapeutic hypothermia (TH) to reduce elevated ICP, and improve outcomes, as assessed by the Glasgow Outcome Scale Extended (GOSE) at 6 months. This was a pragmatic, multi-centre, randomised and controlled trial (1), that showed functional outcomes were worse in the hypothermia group (2). The purpose of this study is to examine possible imbalances in co-interventions that might explain these results.
Methods: An amendment for additional data collection was submitted to the Research-Ethics-Committees and a favourable opinion given in November 2015. All recruiting sites were contacted on the 4
th December 2015. Data gathered on a specialised CRF from January 2016 included: TH induction, treatment of shivering, cause of death, follow-up CT-Brain scan, surgical and medical ICP treatments and infections. For all data collected, descriptive statistics by allocated treatment and dichotomised by GOSE outcome (favourable vs unfavourable) will be generated and further statistical analysis documented.
Results: We anticipate analysing data on up to half the recruited participants (195/387). The outcome of this analysis will be presented for the first time at the IHS 6 meeting.
Conclusion: The pragmatic design of the Eurotherm3235Trial ensured efficient recruitment and data collection. A limitation is the inability to explain unexpected outcomes due to the lack of data on Stage II ICP therapy. We hope to generate a hypothesis to support or refute a biological effect of TH.
Association Between Blood Glucose Levels the Next Day After Targeted Temperature Initiation and Outcome in Traumatic Brain Injury: A Post-Hoc Analysis of the B-HYPO Study
KobataHitoshi1SugieAkira1SuehiroEiichi2DohiKenji3KanekoTadashi4OdaYasutaka5KurodaYasuhiro6YamashitaSusumu7MaekawaTakeshi8B-HYPO Study Group
Osaka Mishima Emergency Critical Care Center
Department of Neurosurgery, Yamaguchi University
Department of Emergency Medicine, The Jikei University
Emergency and General Medicine, Kumamoto University
Advanced Medical Emergency and Critical Care Center, Yamaguchi University
Department of Emergency Medicine, Kagawa University
Emergency and Critical Care Center, Tokuyama Central Hospital
Yamaguchi Grand Medical Center
Background: We investigated associations between blood glucose levels and clinical outcomes in participants of the multicenter prospective randomized controlled Brain-Hypothermia (B-HYPO) study.
Methods: Patients with severe traumatic brain injury (TBI, Glasgow Coma Scale 4–8) were assigned to therapeutic hypothermia (TH, 32–34°C, n = 98) or fever control (35.5–37.0°C, n = 50) groups. TH patients were cooled as soon as possible for ≥72 h and rewarmed at a rate of <1°C/day. We recorded blood glucose (BG) levels on days 0, 1, and 3 after treatment initiation, and day 1 after rewarming. The Glasgow Outcome Scale was assessed at 6 months.
Results: Median BG levels decreased from day 0 to day 1 (163 vs. 132 mg/dL, p = 0.0062) in the fever control group. In contrast, a decrease was observed from day 1 to day 3 (157.5 vs. 126 mg/dL, p < 0.001) in the TH group. Day 1 BG was higher in the TH group compared with the fever control group (p = 0.0252). At day 0, BG levels were higher in non-survivors compared with survivors across all patients (p = 0.0035), the TH group (p = 0.0125), and the non-surgical group (p = 0.0236). Higher day 1 BG levels were observed in non-survivors compared with survivors across all patients (p = 0.0071), the fever control group (p = 0.0495), and the surgical group (p = 0.0364).
Conclusion: In the TH group, the initial stress hyperglycemia was sustained the next day after TH induction. Day 1 BG predicted outcome in TBI patients with TH and fever control. Optimal BG control during fever management may affect outcome and thus should be carefully managed.
Valproate Total Serum Concentrations Misleading When Treating Seizures After Cardiac Arrest
RikerRichard R.MD1GagnonDavid J.PharmD1FraserGilles L.PharmD1MayTeresaDO1SederDavid B.MD1
Tufts University School of Medicine and Maine Medical Center, Portland, ME
Background: Valproate is recommended to treat seizures after cardiac arrest, is highly bound to albumin, and its free fraction can vary due to altered protein binding. We present 4 consecutive patients when total valproate concentrations were misleading and could not be used to predict the free, biologically active, concentration.
Methods: Data from 4 adult patients with myoclonus or seizures during targeted temperature management after cardiac arrest treated with valproate were prospectively evaluated.
Results: Case 1 had persistently low total valproate concentrations despite dose escalations. Simultaneous valproate concentrations revealed a low total concentration (22 mg/L–reference 50–100 mg/L), an appropriate free concentration (19 mg/L– reference 5–25 mg/L) and elevated free fraction (86%) likely related to hypoalbuminemia and propofol administration. Case 2 also had a low total valproate concentration, prompting dose increases. Simultaneous valproate concentrations revealed an elevated free fraction (38%), likely due to uremia and hypoalbuminemia. Case 3 had a total valproate concentration in the reference range, but elevated free concentration and free fraction (52%), likely due to hypoalbuminemia and propofol administration. Case 4 had a low total concentration (25 mg/L) prompting increased valproate dosing, but repeat testing showed an appropriate free concentration (18 mg/L) despite a low total (42 mg/L). All 4 cases developed hyperammonemia.
Conclusion: The free fraction of valproate may increase with hypoalbuminemia, uremia, concentrations in the upper reference range, and substances that displace valproate from albumin (e.g., aspirin, propofol, and free fatty acids). Routine monitoring of free valproate concentrations might prevent inappropriate dose escalations and associated toxicity.
Validation of Suppression Ratio from Simplified and Full Montage EEG During Targeted Temperature Management After Cardiac Arrest
RikerRichard R.MD1CraigAlexaMD1EubankLouis1MayTeresaDO1SederDavid B.MD1
Tufts University School of Medicine and Maine Medical Center, Portland, ME
Background: The Suppression Ratio (SR), a processed electroencephalographic (EEG) variable estimating the degree of suppression, has been associated with neurological outcome after acute brain injury from multiple causes. Different technical methods are used to calculate SR. We tested the construct validity of SR during simultaneous simplified and full EEG montage recordings during targeted temperature management (TTM).
Methods: A convenience sample of adult patients treated with TTM after cardiac arrest were evaluated to compare the SR determined by the Medtronic bispectral index monitor (MSR) and the full montage continuous EEG using Natus equipment with Persyst Magic-Marker software (PSR). A Neurologist board certified in epilepsy scored the full montage amplitude integrated EEG (aEEG) pattern and background activity using 2012 standardized ACNS terminology blinded to SR. SR scores were compared for the various EEG categories using Kruskal-Wallis ANOVA.
Results: 22 adults were evaluated with a median age of 54 years, 14 (64%) were male. During SR collection, core temperature was 33.8°C and patients were receiving propofol at 20 mcg/kg/min and fentanyl at 37.5 mcg/hr. The MSR varied from 0% when background was continuous or nearly continuous, to 23% when discontinuous with suppression, to 64% during burst suppression (p = 0.010). The MSR was 46% during flat aEEG, 34% during aEEG Burst Suppression, and 0% during aEEG continuous patterns (p < 0.001). The PSR showed similar differences.
Conclusion: The Suppression Ratio calculated by Medtronic simplified EEG monitors and Natus-Persyst full montage monitors is a valid measure of differing amplitude integrated EEG patterns and EEG background classifications.
Neurobiological Effect of Selective Brain Cooling After Concussive Injury
WalterAlexa1JohnsonBrian1FinelliKatie1BaiXiaoxiao1TessadaJohn2SlobounovSemyon1
The Pennsylvania State University, University Park, PA
Spartan Medical Inc., Silver Spring, MD
Background: The use of hypothermia in treating traumatic brain injury has been previously demonstrated to have positive patient outcomes; however, selective brain cooling has not been well demonstrated in a concussed population. Therefore, we investigated the effect of selective cooling of the head and neck in collegiate athletes on symptom resolution (via virtual reality (VR)) and physiological function (via MRI) in the acute phase of injury.
Methods: Subjects were divided into two groups: control (NV; no previous concussion) and concussion (CN; ≤10 days of injury). Subjects underwent a MRI scan, including sequences of resting state functional MRI (rsfMRI) and arterial spin labeling (ASL), then a 30 minute cooling period (at 5°C) followed by the same MRI sequence.
Results: rsfMRI had no significant differences between pre- and post-cooling in the default mode network for either group; however, within CN group, visually the number and strength of connections was reduced in the post-cooling scan. ASL sequences for CN group revealed a significant increase (p < 0.05) in relative cerebral blood flow (relCBF) in cortical and subcortical cortex post-cooling while NV group had significantly decreased (p < 0.05) relCBF post-cooling. For VR, half of CN subjects could not tolerate it due to reoccurrence of symptoms, but those who did complete testing had improved scores in spatial navigation (p ∼ 0.09) and reaction time (p ∼ 0.07) post-cooling.
Conclusion: These results suggest that 30 minutes of selective cooling can induce positive physiological and neurocognitive changes in a concussed population which could lead to its future use as a potential treatment modality.
Using Invasive Monitoring of Intracranial Physiology to Guide Targeted Temperature Management After Hypoxic Ischemic Brain Injury
RajagopalanSwarnaMD1BaluRamaniMD1BakerWesleyPhD1LevineJoshuaMD1AbellaBenjaminMD1KofkeAndrewMD1
University of Pennsylvania
Background: Brain damage after global hypoxic/ischemic insults is not instantaneous, but rather evolves over time. Targeted temperature management (TTM) can reduce delayed secondary brain injury (SBI); however, the mechanisms that produce neuroprotection with TTM and the optimum duration and intensity of temperature control are unknown. We used invasive intracranial Multimodality Monitoring (MMM) to investigate the occurrence of indicators of SBI during Targeted Temperature Management (TTM) after hypoxic ischemic brain injury (HBI).
Methods: Retrospective analysis of 3 patients with HIBI cared for between November 2015 to May 2016 that underwent MMM to guide-post resuscitation care. MMM consisted of an ICP monitor, brain oxygen (PbtO2) monitor, cerebral blood flow (CBF) probe in all patients, and in addition included a microdialysis probe in the other patient.
Results: Etiology of HIBI included cardiac arrest (2 patients) and prolonged hypoxemia due to hanging (1 patient). Target temperature was 33°C in 2 patients and 36°C in 1 patient. The median duration of TTM was 79 hours. Both patients that died had greater than ten ICP crises associated with low PbtO2 over 50 hours, ultimately becoming refractory to treatment in their final 30 hours before death. LPR elevation preceded refractory ICP crisis in the patient with microdialysis. The duration of TTM was extended based on MMM data in one patient due to persistent ICP crisis when rewarming was attempted.
Conclusions: Knowledge of disturbances in intracranial physiology after cardiac arrest and TTM is limited. Intracerebral MMM can offer valuable information about cerebral physiology and guide therapy.
Early Evaluation of Post-Arrest Injury and Tailored TTM
KurodaYasuhiro1KawakitaKenya1HifumiToru1TakanoKoshiro1AbeYuko1ShinoharaNatsuyo1HamayaHideyuki1OkazakiTomoya1ManabeArisa1HagiikeMasanobu1
Department of Emergency, Disaster, and Critical Care Medicine, Faculty of Medicine, Kagawa University
Background: Brain damage after return of spontaneous circulation (ROSC) varies among studies and patients despite their comatose status (i.e., lack of meaningful response to verbal commands), because of the absence of an established modality enabling proper evaluation. Evaluation of brain injury after ROSC is needed for the determination of the inclusion criteria of neurocritical care, especially of targeted temperature management.
Methods: The association between admission Glasgow Coma Score (GCS) motor score and neurologic outcome remains unknown in comatose cardiac arrest survivors. We found that GCS motor scores immediately after ROSC (day 0) is an independent predictor of good neurologic outcome at 90 days in patients sustaining out-of-hospital cardiac arrest who receive therapeutic hypothermia: GCS motor score 1, n = 130 (52.2%); score 2–3, n = 23 (76.7%); score 4–5, n = 20 (87.0%), P < 0.01 (Hifumi 2015 2201).
Results: Recently no significant differences of neurologic outcome at 30 days after hospital admission was observed between mild therapeutic hypothermia and control in the subgroup of GCS Motor score 5 or 6. These data show that initial GCS motor score examination immediately after ROSC can at least provide baseline objective prognostic data for decisions by healthcare professionals.
Conclusion: Neurological signs such as GCS, brain stem reflex, respiratory status, and degree of shivering are potential variables that can be incorporated into a predictive model for a more precise evaluation of brain injury in cardiac arrest survivors undergoing TTM. Effect of targeted temperature management should be evaluated depending on the brain injury in PCAS.
Predicting Fever Post-TTM in Intracerebral Hemorrhage
MukhtarUmerMBBS1BoormanDavidMD1HarshyneLarryPhD1HooperD. CraigPhD1RinconFredMD1
Thomas Jefferson University
Background: The objective of this study was to determine if post-TTM fever would be predicted by changes in chemokine/cytokine profiles in ICH patients undergoing TTM.
Methods: During a Phase-II clinical trial of temperature modulation after ICH, (TTM-ICH, clinicaltrials.gov: NCT01607151), we measured cumulative levels of TNF-α, IL-1, IL-6 and IL10 in blood of enrolled patients. Patients were randomized to moderate hypothermia (MH, core temp [Tc] 33–34°C) or normothermia (NT, Tc 36–37°C). Blood was collected daily from admission day (D0) till day 7 (D7). Post-TTM, we measured the temperature (T°C) four times a day from D8-D15, and ascertained the incidence of post-TTM fever, which was defined as any T-max ≥38.3°C. Chemokine/cytokine levels were log-transformed. We modeled the effect of cytokine/chemokine and TTM group on the incidence of fever using Generalized Estimating Equations (GEE). Adjusted for multiple comparisons (one GEE model per chemokine), an alpha (α) level of <.0125 or 0.05 ÷ 4 was considered significant.
Results: In total, 9 ICH patients participated in this study. Median age was 57 years (IQR 22), 55% women, 66% Blacks, median GCS was 7 (IQR 4), median ICH score was 2 (IQR 1), median NIHSS 20 (IQR 5). Adjusted for treatment group, higher logTNF-α (Model 1 OR, 3118; 95%CI, 159–61464, p < .0001) and logIL-1 (Model 2 OR, 9.4; 95%CI, 1.9–46, p = 0.006) predicted post TTM fever. Treatment with MH predicted lower incidence of post-TTM fever (Model 1 OR, 2×10−5; 95%CI, 4×10−8-.013-46, p = 0.001).
Conclusions: Treatment with normothermia and higher logTNF-α and logIL1 predicted post-TTM fever in ICH patients. Our results deserve further study.
Targeted Temperature Maintenance Using an Esophageal Heat Transfer Device
NaimanMelissaPhD1MarkotaAndrejMD2HegazyAhmedMB BCh, FRCPC3DingleyJohnMD4KulstadErikMD5
University of Illinois at Chicago
University Medical Center Maribor
University of Western Ontario
Welsh Centre for Burns
Advocate Christ Medical Center
Background: Preventing unintentional hypothermia, reducing fever, and inducing therapeutic hypothermia when appropriate, are each tied to positive health outcomes. While current evidence does not support the superiority of a particular TTM method, clinical data do suggest that target temperature maintenance is a key component of successful TTM. This study sought to determine the ability of an esophageal heat transfer device (EHTD) to maintain target temperature in critical care cases.
Methods: De-identified data for subjects who received temperature management using an EHTD were collected from three clinical sites. All patients were at least 18 years of age, weighed at least 40 kg, and presented with a condition appropriate for TTM. Core temperature readings for each patient were recorded at least hourly; if measurements were recorded more frequently, temperature over an hour span was averaged. Patient data was analyzed to determine how many measurements were within ±0.5°C or ±1.0°C of the target, starting with the first measurement within range and up to 23 consecutive readings thereafter.
Results: Data from 30 patients were analyzed, including 23 post-cardiac arrest patients receiving targeted temperature management, 4 fever reversal cases, and 3 burn patients receiving perioperative normothermia. A total of 642 core temperature measurement events were included in the analysis. 546 (85%) were within ±0.5°C the designated target range and 619 (96%) were within ±1.0°C.
Conclusion: The EHTD successfully maintained target temperature in a variety of TTM protocols. Secondary analysis also revealed that in post-cardiac arrest cases, 99.3% of readings remained below 38°C during rewarming.
Time to Chill: Revising and Sustaining a Targeted Temperature Management Program in a Community Hospital
BoydLindsayRN, MS1
Overlake Hospital Medical Center
Background: The implementation and sustainability of a targeted temperature management program at a community hospital can be challenging secondary to factors such as resources and a smaller number of patients receiving the therapy.
Methods: Policy and order set was revised to reflect current practice. A plan for ongoing education of all stakeholders was implemented. This included both didactic and hands on sessions. Stakeholders from all departments that care for these patients were included. Data was collected from cases and shared with staff to highlight areas that went well and any areas that required improvement.
Results: An improvement of door/arrest to goal temperature times was noted. A decrease in number of “missed” opportunities to implement the therapy was achieved. Staff reported an increase comfort level in implementing both hypothermia and the additional component of normothermia therapies.
Conclusion: A continued focus is required to support an effective targeted temperature management in a community hospital. Sharing data from real cases back with all key stakeholders is key in increasing buy in and to focus on areas that require improvement. An increased awareness and knowledge of the therapy across all staff helps decrease the time taken to implement this important therapy.
A Novel Approach to Quality Improvement in a Therapeutic Hypothermia Program for Neonates with Hypoxic Ischemic Encephalopathy
ArmstrongErinBS, RN, CCRN1RitterJaimeMPH, CCRP1LawrenceAmyBSN, RN, CCRN1
C. R. Bard, Inc., Bard Medical Division
Background: Studies indicate that therapeutic hypothermia (TH) is a safe and effective neural rescue strategy for infants who have clinical evidence of moderate/severe neonatal encephalopathy. Hospitals treating these patients have established protocols for providing TH. We used performance metrics to assess compliance with one such protocol and to identify areas for improvement.
Methods: Data was downloaded directly from a targeted temperature management (TTM) device for 14 patients treated with TH for neonatal encephalopathy. This data was used to calculate performance metrics such as birth to target temperature time and time at target temperature and to compare these metrics to the hospital's protocol goals. Product training verification was performed during unit rounds to assess the level of clinician knowledge regarding the proper use of the TTM device.
Results: The birth to start protocol goal was within 6 hours of birth; actual mean time was 3:44 hours. The time to target temperature (induction phase) protocol goal was within 4 hours from initiation; actual mean time was 2:17 hours. The time at target temperature (maintenance phase) protocol goal was 72 hours; actual mean time was 69:52. The rewarming protocol goal was 6 hours; actual mean time was 8:25 hours. The product training verification assessment revealed knowledge gaps around a variety of topics including interrupting, completing, and managing therapy.
Conclusion: Information downloaded directly from a TTM device and competency verification provides a novel approach to assessing performance metrics for TH. While overall compliance with the hospital's protocol was good, product knowledge gaps were identified.
Novel Esophageal Cooling Device for Therapeutic Normothermia
BadjatiaNeeraj1KhanImad1HaymoreJoseph1MelinoskyChristopher1BautistaMary Ann1ChangWan-Tsu1ParikhGunjan1MottaMelissa1
University of Maryland
Background: Achieving and maintaining normothermia (NT) after subarachnoid hemorrhage (SAH) or intracerebral hemorrhage (ICH) often requires surface or intravascular cooling devices that are associated with a significant burden of shivering. We describe a new, closed loop esophageal cooling device (ECD: Esophageal Cooling Device; Advanced Cooling Therapy: Chicago, IL) connected to a Stryker Medi-Therm (Stryker Corporation: Kalamazoo, MI) system to induce NT (37° ± 0.5° C) and the shiver burden during the maintenance of NT.
Methods: We enrolled mechanically ventilated patients with SAH or ICH with refractory fever (>38.3°C). Temperature and Bedside Shivering Assessment Scale (BSAS) were recorded every 15 minutes for the first 2 hours, then hourly. Success and time to NT, hourly temperature burden (TB: >37.5°C/hr.), percent time above 38°C, median BSAS and cumulative number of anti-shivering interventions per patient was recorded prospectively. All patients received magnesium, buspirone, and acetaminophen as baseline anti-shivering interventions.
Results: Ten patients (7 ICH, 3 SAH) were enrolled between October 2015 and April 2016. The median GCS at initiation was 6 (4–11), age 52 ± 6 years old, BMI 25 ± 6 kg/m2, BSA 1.8 ± 0.2 m2, and 70% were women. There was a temperature reduction at 120 minutes (mean 38.7 C to 37.9 C, p = 0.005) and 90% of patients achieved NT (median time = 4.5 hrs.; range: 0.5 – 38 hours). NT was maintained for median 91 hours (range: 15 to 131 hours) with a TB of 0.05 ± 0.5 C*hr, and 15% (range: 0 – 28%) time above >38.0°C/patient. Median BSAS = 0 with any shiver (BSAS >0) occurring 7.6% of the time. The median number of total shiver interventions per patient was 5 (1 – 22) throughout the TTM time period. No device related complications were noted.
Conclusions: The ECD successfully achieved and maintained NT with a low burden of shivering and may be a feasible option for NT in this critically ill population.
Rewarming for Refractory Accidental Hypothermia Using Extracorporeal Membrane Oxygenation
HiroseHitoshiMD1CavarocchiNicholasMD1
Thomas Jefferson University Hospital
Background: Accidental hypothermia complicated by cardiac arrest carries a high mortality rate in urban areas. For moderate hypothermia cases conventional rewarming methods are usually adequate, however in severe cases extracorporeal membrane oxygenation (ECMO) is known to provide the most efficient rewarming with complete cardiopulmonary support. We report a case of severe hypothermia complicated by prolonged cardiac arrest successfully resuscitated using ECMO.
Case Report: A 45 year old female was brought to our emergency department with a core body temperature <25°C. Shortly after arrival she had witnessed cardiac arrest in the department. Resuscitative efforts were started immediately including conventional rewarming techniques, followed by ECMO support. ECMO was used successfully in this case to resuscitate this patient from prolonged arrest (3.5 hours) when conventional techniques likely would have failed. After a prolonged hospital course this patient was discharged with her baseline mental and physical capacities intact.
Conclusions: This case demonstrates the advantages of advanced internal rewarming techniques, such as ECMO, for quick and efficient rewarming of severely hypothermic patients. This case supports the use of ECMO in severely hypothermic patients as the standard of care.
Extracorporeal Corporeal Cardiopulmonary Bypass Under Cardioplumonary Resuscitation
HiroseHitoshiMD1PitcherHarrison T.MD1YangQiongMD1CavarocchiNicholasMD1
Thomas Jefferson University Hospital, Philadelphia, PA
Introduction: Despite advances in medical care, survival to discharge and full neurological recovery after cardiac arrest (CA) remains <20%. An alternate approach to traditional CPR is E-CPR, which provides immediate cardiovascular support when traditional methods fail. Renewed interest in ECMO has resulted in the use of ECMO during CPR (E-CPR) to improve outcomes.
Methods: Between 2010 and 2013, a total of 100 ECMO procedures were performed at our institution. 24 cases unresponsive to conventional CPR had E-CPR. Patient demographics, survival to discharge, and neurological recovery were retrospectively analyzed with IRB approval.
Results: Of the 24 patients who received E-CPR, there were 15 males and 9 females, with a mean age of 47 ± 15 years. The etiologies (#) for E-CPR in these patients were: acute myocardial infarction (9), malignant arrhythmia (4), myocarditis (3), acute pulmonary emboli (2), hypothermia (2), and 1 case each of acute rejection, drug overdose induced cardiac arrest, post-cardiotomy failure, and septic shock. The mean duration of ECMO support was 5.1 ± 4.7 days. All patients who met criteria were placed on 24 hr. hypothermia protocol with initiation of ECMO. 13 of 24 (54%) patients survived E-CPR. 7/13 patients were discharged without any neurological sequence. 6/13 patients died post-ECMO, but pre-discharge (anoxic brain injury (4), sepsis (1), stroke (1)). The causes of the death on E-CPR were: anoxic brain injury (5), stroke (3), metabolic acidosis (1), bowel necrosis (1), and family's withdrawal (1). 2/5 patients with anoxic brain injury donated multiple organs for transplant while on E-CPR. The hospital discharge to survival rate was 54% (7/13 patients) with full neurological recovery.
Conclusions: The E-CPR provided improved survival and neurological recovery compared to national in-hospital post-CPR statistics. E-CPR also made multi-organ procurement possible. The protection of patients' brains remains an issue to be addressed in order for survival rates to be further improved.
Temperature Changes in the Porcine Brain During High Flow Cold Air Exposure: First MRI Measurements
KjorstadA.1FiehlerJ.1SedlacikJ.1TemmeF.1
Neuroradiology, University Medical Center Hamburg-Eppendorf, Hamburg, Germany
Background: As of yet, no consensus has been reached on the best and most cost-effective protocol for inducing hypothermia. We propose and demonstrate the cooling properties of a protocol using high flow cold air applied directly to the nasal cavity of pigs, utilizing the close proximity of the nasal airways and the brain. Temperature changes are measured in vivo using an established MRI temperature technique.
Methods: 5 adult pigs were anesthetized and put under mechanical ventilation. A cold air compressor (MECOTEC cryoAir) delivering cold air measured at −10° C and flow above 160 liter/min was used for the cooling. A small tube connected to the compressor was then inserted about 1 cm into the snout of the pigs. The in vivo brain temperature change was then measured during cooling using a MRI phase temperature technique utilizing the temperature dependence of the proton resonance frequency.
Results: A significant (p < 0.05) cooling effect was seen in the anterior parts of the brains. Averaging over time and all pigs a nonsignificant whole brain cooling effect of −0.33 ± 0.30°C was found. The anterior part, being directly exposed to the cold air, showed a larger and significant effect, with an average change of −0.83 ± 0.54°C.
Conclusion: Small, but significant cooling effects can be achieved locally in the anterior parts of porcine brains using a high flow cold air protocol. The protocol is fast and easy to administrate and could potentially be used by first responders if translatable to humans. The first human tests are in progress.
Accuracy of Predictive Basal Metabolic Rate Equations and the Safety of Enteral Nutrition During TTM in ICH Patients
DobakStephanieRD, MS1RinconFredMD, MSc1
Thomas Jefferson University
Background: Though early enteral nutrition (EN) is associated with improved outcomes, it is often deferred until the rewarming phase post-TTM. We sought to determine the accuracy of predictive basal metabolic rate (BMR) equations and the safety and tolerance of EN during TTM after ICH.
Methods: Patients were randomized to 72 hours of moderate hypothermia (MH; Tcore33–34°C) or normothermia (NT; Tcore36–37°C). Harris-Benedict (BMR-HB) and Penn-State equation (BMR-PS) calculations were compared to indirect calorimetry (IC) at Day(D)0 and D1-3. MH patients received trophic semi-elemental gastric EN. Occurrences of feeding intolerance, GI-related adverse events and ventilator-associated pneumonia (VAP) were analyzed with double-sided matched pairs t-test.
Results: 13 ICH patients participated: 6 MH, 7 NT. Mean time to initiate EN: 29.9(MH) vs 18.4(NT) hours (p = 0.046). Average daily EN calories received D0-3: 398(MH) vs 1006(NT) (p < 0.01). Three MH patients experienced high gastric residuals prior to prokinetic agent use, 1 mild ileus; 1 NT patient vomited. No GI-related adverse events were reported. One MH and 1 NT patient had VAP. Two MH patients received IC, and from D0 to D1-3 BMR-HB remained stable (1331 kcal); BMR-PS decreased (1511 kcal vs 1145 kcal, p = 0.5); IC decreased (1413 kcal vs 985 kcal, p = 0.2).
Conclusions: In ICH patients undergoing MH, REE is decreased and predictive equations overestimate BMR. EN is feasible, though delayed EN initiation, high gastric residuals, and less EN provision are common.
