Impact of intra-arrest therapeutic hypothermia in outcomes of prehospital cardiac arrest: a randomized controlled trial

Transcription

1 Intensive Care Med DOI /s x SEVEN-DAY PROFILE PUBLICATION Guillaume Debaty Maxime Maignan Dominique Savary François-xavier Koch Stéphane Ruckly Michel Durand Julien Picard Christophe Escallier Renaud Chouquer Charles Santre Clemence Minet Dorra Guergour Laure Hammer Hélène Bouvaist Loic Belle Christophe Adrie Jean-François Payen Françoise Carpentier Pierre-Yves Gueugniaud Vincent Danel Jean-François Timsit Impact of intra-arrest therapeutic hypothermia in outcomes of prehospital cardiac arrest: a randomized controlled trial Received: 8 September 2014 Accepted: 9 October 2014 Ó Springer-Verlag Berlin Heidelberg and ESICM 2014 Electronic supplementary material The online version of this article (doi: /s x) contains supplementary material, which is available to authorized users. G. Debaty M. Maignan UJF-Grenoble 1/CNRS/TIMC-IMAG UMR 5525/Team PRETA, Grenoble, France G. Debaty ()) M. Maignan F. Koch C. Escallier F. Carpentier V. Danel SAMU 38, Pôle Urgences-Médecine Aiguë, CHU Grenoble, 10217, Grenoble Cedex 09, France gdebaty@chu-grenoble.fr Tel.: (?33) D. Savary Department of Emergency Medicine, SAMU 74, Annecy Hospital, Annecy, France S. Ruckly J.-F. Timsit University Joseph Fourrier, Albert Bonniot Institute, La Tronche Cedex, France M. Durand J. Picard J.-F. Payen Department of Anesthesiology and Critical Care, Grenoble University Hospital, Grenoble, France R. Chouquer C. Santre Department of Anesthesiology and Critical Care, Annecy Hospital, Annecy, France C. Minet L. Hammer Department of Intensive Care Medicine, Grenoble University Hospital, Grenoble, France D. Guergour Department of Biochemistry Toxicology Pharmacology, Grenoble University Hospital, Grenoble, France H. Bouvaist Department of Cardiology, Grenoble University Hospital, Grenoble, France L. Belle Department of Cardiology, Annecy Hospital, Annecy, France C. Adrie Physiology Department, Cochin Hospital, ICU, Paris, France S. Ruckly P.-Y. Gueugniaud SAMU 69, Hospices Civils de Lyon, University of Lyon 1, Lyon, France J.-F. Timsit UMR 1137-IAME Team 5-(DeSCID) Decision Sciences in Infectious Diseases, Control and Care, Inserm/Univ Paris Diderot, Sorbonne Paris Cité, Paris, France Abstract Purpose: Mild therapeutic hypothermia (TH) is recommended as soon as possible after the return of spontaneous circulation to improve outcomes after outof-hospital cardiac arrest (OHCA). Preclinical data suggest that the benefit of TH could be increased if treatment is started during cardiac arrest. We aimed to study the impact of intra-arrest therapeutic hypothermia (IATH) on neurological injury and inflammation following OHCA. Methods: We conducted a 1:1 randomized, multicenter study in three prehospital emergency medical services and four critical care units in France. OHCA patients, irrespective of the initial rhythm, received either an infusion of cold saline and external cooling during cardiac arrest (IATH group) or TH started after hospital admission (hospital-cooling group). The primary endpoint was neuron-

2 specific enolase (NSE) serum concentrations at 24 h. Secondary endpoints included IL-6, IL-8, and IL-10 concentrations, and clinical outcome. Results: Of the 245 patients included, 123 were analyzed in the IATH group and 122 in the hospital-cooling group. IATH decreased time to reach temperature B34 C by 75 min (95 % CI: 4; 269). The rate of patients admitted alive to hospital was not different between groups [IATH n = 41 (33 %) vs. hospital cooling n = 36 (30 %); p = 0.51]. Levels of NSE and inflammatory biomarkers were not different between groups [median NSE at 24 h: IATH 96.7 lg/l (IQR: ) vs. hospital cooling 97.6 lg/l (IQR: ), p = 0.64]. No difference in survival and cerebral performance were found at 1 month. Conclusions: IATH did not affect biological markers of inflammation or brain damage or clinical outcome. Keywords Therapeutic hypothermia Cardiopulmonary resuscitation Resuscitation Cardiac arrest Post-cardiac arrest syndrome Introduction Approximately 300,000 patients suffer from out-of-hospital cardiac arrest (OHCA) annually in Europe, and survival rate with favorable neurological function remains dismal [1]. Mild therapeutic hypothermia (TH) has been shown to improve survival and reduce neurological damage by limiting reperfusion injury after OHCA [2]. TH (32 34 C) is recommended for h as soon as possible after return of spontaneous circulation (ROSC) [3]. Therefore, the optimal timing for induction of TH and the optimal temperature remain uncertain [4]. In most studies, TH is induced after hospital admission, and time to reach the targeted temperature can be up to 8 h [2, 5]. Some clinical studies have shown that inducing TH in the prehospital setting is feasible. While the prehospital use of external ice packs is limited by a low cooling rate, the infusion of a large volume (30 ml/kg) of 4 C has been shown to decrease temperature between 1.2 and 2.3 C before hospital admission in OHCA patients [6 9]. However, prehospital cooling when performed after ROSC seems to be ineffective to improve survival or neurological outcomes [7 9]. This lack of benefit could be linked to the delay in the initiation of cooling. Several animal studies and some clinical data have suggested that the benefit to neurological outcome could be greatly increased if TH is started during cardiopulmonary resuscitation (CPR) [10 14]. Experimental investigations have also shown that intra-arrest therapeutic hypothermia (IATH) increases the success rate of defibrillation attempts in ventricular fibrillation (VF), improved left ventricular function, and reduced myocardial infarct size [15, 16]. Serum biochemical markers and inflammatory response markers, such as neuron-specific enolase (NSE), S100 calcium binding protein B (S-100B), and interleukin (IL)-6, -8, and -10, have been proposed as early prognostic markers after successful resuscitation of cardiac arrest [17 19]. Specifically, NSE has been reported to be the most promising and extensively studied marker to predict bad outcome after cardiac arrest [3]. Elevated level of NSE between 24 and 72 h were associated with initial illness severity [20]. The aim of the present study was to investigate the effects of IATH on neurological outcome and biological markers after OHCA. We hypothesized that IATH would be associated with lower neurological damage and reduced serum levels of NSE. Methods Study design The study was approved by the institutional review board (CPP Sud Est V, 01/2009, Lyon, France). In accordance with French law, the board waived the requirement for obtaining informed consent from patients because of the emergency setting of the research; however, the subject s legal representative was informed of the subject s study participation, and patients who regained normal neurological function were asked to provide their consent for use of the data. The study was designed as a 1:1 randomized, multicenter trial. Three prehospital emergency medical services (EMS) and four intensive care units in two hospitals in France participated in the study from September 2009 to July In France, prehospital EMS consist of mobile intensive care units staffed by a paramedic driver, a nurse, and a senior emergency physician. Mobile intensive care units are dispatched in case of OHCA to provide advanced life support in addition to first rescuers response. As a consequence, patients are admitted to hospital only if they had a ROSC. In this study, OHCA patients over 18 years of age were recruited during resuscitation by the EMS physician. Patients eligible for advanced life support were included irrespective of rhythm. Patients with trauma, hemorrhage, asphyxia, already hypothermic (temperature \34 C), women who were obviously pregnant, patients who had achieved ROSC before randomization, and patients with a do-not-attempt resuscitation order were not included. The

3 randomization assignments were generated under a randomized permuted-block design, with block sizes of 4, in a 1:1 allocation. Each mobile intensive care unit was given sequentially numbered sealed envelopes containing single randomization assignments. The randomization list was stratified by centers. The envelope was opened if the patient fulfilled the inclusion criteria, and the patient was assigned to either IATH or to cooling treatment once admitted to the hospital (hospital cooling group). Treatment protocol In all patients, the resuscitation attempt followed the European Resuscitation Council guidelines [1]. The 2010 changes in the guidelines have been implemented in all centers. At mobile intensive care unit arrival, intravenous access was established. Concomitantly, airways were secured by endotracheal intubation and core temperature was measured using an esophageal probe. In addition to usual care, IATH patients received an infusion of up to 2,000 ml of ice-cold 0.9 % saline solution at 100 ml/min during cardiac arrest by use of a standard infusion set and a pressure bag inflated to 300 mm Hg. Surface cooling was also induced using gel pads (Colpac; Chattanooga Medical Supply). Four cooling units were applied on the neck, abdomen and thighs. Fluid and surface cooling supplies were stored in an insulated container to ensure that temperature was\8 C. In the IATH group, core temperature was maintained between 32 and 34 C. Following ROSC, intravenous midazolam (0.1 mg/kg), sufentanil (0.2 lg/ kg) and cisatracurium (10 mg) were administered through continuous intravenous infusion to suppress shivering. Following hospital admission, TH was either maintained (IATH group) or induced (hospital cooling group) using cold saline infusion, cooling mattress, cold air circulation and/or extra corporeal life support if needed for hemodynamic support. Cooling was induced at a maximum rate with a target temperature of C. In both groups, the target temperature was maintained for 24 h, and a controlled rewarming phase was started at a set rate of C per hour to reach 37 C. ROSC was defined as an organized rhythm and palpable pulse that was sustained for at least 2 min. All subjects received standard post-resuscitation treatment on hospital arrival per the institution s standards. Intravenous sedation, analgesia, and neuromuscular blockade were initiated or continued. Ventilation was set to maintain normoxemia as well as normocapnia. Secondary endpoints included IL-6, IL-8 and IL-10 concentrations during the first 72 h, cooling rates, ROSC rate, length of stay in the intensive care, survival (discharge, 30 days and 1 years), and neurological outcome (hospital discharge and 30 days). Secondary endpoints were also analyzed according to initial rhythm. NSE, IL- 6, IL-8 and IL-10 have been proposed as early prognostic markers after successful resuscitation of CA and have been evaluated in patients treated with hypothermia, in particular NSE has been the only marker recommended in American Guidelines in 2006 to predict outcome after cardiac arrest [17 19, 21, 22]. A safety end point was defined as the rate of complications during 72 h after OHCA. Complications included the rate of pulmonary edema, bleeding (bleeding causing fatality, intracranial bleeding, and other bleeding that required transfusion), infection (sepsis, pneumonia, urinary tract infection) with positive culture, arrhythmia (VF, ventricular tachycardia, atrial flutter, atrial fibrillation, need for pacing, new cardiac arrest), seizures, and hyperthermia ([38 C within 72 h). Data collection Data on cardiac arrest for individual patients were recorded in the Utstein style [23]. NSE, IL-6, IL-8, and IL-10 were collected and analyzed at baseline (just after randomization in the prehospital setting), and 3, 6, 12, 24, 48 and 72 h after baseline. Blood samples were stored on ice and centrifuged as soon as possible for 15 min at 2,000g at 4 C, and plasma was stored at -80 C until batch-wise analysis. Measurement of serum NSE was performed using an enzyme immunoassay by an automated Elecsys (Roche Diagnostics, Germany). Interleukins were determined using Biochip Array Technology (Randox Laboratories, Antrim, UK). Haemolyzed samples were excluded. Other clinical and biological data were collected at baseline, every 15 min in the prehospital setting, at hospital admission, at ICU admission, and 3, 6, 12, 24, 48 and 72 h after ICU admission. A chest x-ray was performed at admission and at 24 h. A cardiac ultrasound evaluation was performed 24 h after admission. Neurological outcome at hospital discharge and 30 days was assessed by a physician blinded to randomization using the five-grade Glasgow-Pittsburgh cerebral performance categories (CPC) [23]. Outcome The aim of this study was to compare neurological damage and inflammation following OHCA between IATH and standard groups. We chose as primary endpoint the concentration of NSE at 24 h. Statistical analysis The sample size was estimated a priori to measure a difference of NSE of 5 lg/l at 24 h with an estimated mean value of 10 lg/l in the IATH group and 15 lg/l in the hospital cooling group (standard deviation was

4 1559 Pa ents with out-of-hospital cardiac arrest assessed for eligibility 592 cardiac arrests considered for inclusion 967 excluded 393 Dead 108 CPR stopped at ALS team arrival 24 Aged <18 years 111 Trauma 123 Asphyxia 14 Hemorrhage 84 ROSC before randomiza on 110 Others 245 pa ents randomized 347 cardiac arrests not included for hypothermia (< 34 C), randomiza on kit not available or unknown reasons 123 pa ents randomized to Intra-arrest therapeu c hypothermia 122 pa ents randomized to 5 randomized pa ents found to violate inclusion criteria - 4 pa ents with hypothermia (< 34 C) - 1 pa ents with obvious hypoxic e ology of cardiac arrest 7 randomized pa ents found to violate inclusion criteria - 4 pa ents with hypothermia (< 34 C) - 1 pa ent with hemorrhage - 2 pa ents with obvious hypoxic e ology of cardiac arrest 123 Included in inten on-to-treat analysis 122 Included in inten on-to-treat analysis Fig. 1 Flow chart. CPR cardiopulmonary resuscitation, ALS advanced life support, ROSC return of spontaneous circulation Table 1 Baseline characteristics of the patients Characteristics Intra-arrest therapeutic hypothermia (123 patients) (122 patients) Missing value Age, years, median (IQR) 66 (57.9; 75.9) 68.5 (55.6; 77.9) 0 Male sex, n (%) 89 (72.4) 86 (70.5) 0 Witnessed arrest, n (%) 103 (83.7) 91 (74.6) 0 Bystander CPR, n (%) 61 (49.6) 63 (51.6) 0 Cardiac etiology of arrest, n (%) 107 (87) 101 (82.8) 0 Location of arrest, n (%) Home 91 (74) 86 (70.5) 0 Public place 26 (21.1) 20 (16.4) Other 6 (4.9) 16 (13.1) Emergency call to BLS arrival, min, median (IQR) 10 (6, 13) 10 (5, 15) 1 Time from call to ALS arrival, min, median (IQR) 17 (11, 25) 17.5 (13, 25) 0 Time from ALS arrival to ROSC, min, median (IQR) 15.5 (11, 20) 18 (12, 27) 1 Time from collapse to ROSC, min, median (IQR) 27 (20, 35) 30 (20, 36) 1 Initial rhythm, n (%) 0 VF/VT 36 (29.3) 32 (26.2) PEA 8 (6.5) 9 (7.4) Asystole 79 (64.2) 81 (66.4) No. of defibrillations on scene, median (IQR) 0 (0; 3) 0 (0; 3) 0 EtCO2 maximum value during CPR, mmhg, median (IQR) 25 (14.5; 38) 26 (16, 40) 22 Epinephrine during CPR, mg, median (IQR) 7 (5, 10) 7 (5, 10) 1 Amiodarone, mg, median (IQR) 0 (0; 300) 0 (0; 300) 2 CPR cardiopulmonary resuscitation, BLS basic life support, ALS advanced life support, ROSC return of spontaneous circulation, VF/VT ventricular fibrillation/ventricular tachycardia, PEA pulseless electrical activity

5 estimated at 5 lg/l in both groups) [22]. According to a French study, an estimated 21 % rate of survival at hospital admission was expected [24]. Thus, a sample size of 250 patients (or 23 patients per group alive at hospital admission) was required to achieve 80 % power with a a-error of The study was not powered to address clinical outcome. For descriptive statistics, quantitative variables were reported as medians with inter-quartile ranges (IQR) and were compared using the Wilcoxon test. Categorical data were reported as percentages, and compared using the Chi square test. p values \0.05 were considered significant. Missing biological values were recoded from time zero to 72 h using multiple imputation for longitudinal studies using IVEware software (University of Michigan, Ann Harbor, MI, USA) [25, 26]. Linear mixed models with random effect were used to compare the evolution of biomarkers in both groups. One model for each biomarker (NSE, IL-6, IL-8 and IL-10) was computed. The target model was the biomarker value, adjusted by time (3, 6, 12, 24, 48 and 72 h), group (IATH vs. hospital cooling), the first biomarker value (time 0), and an interaction between time and group. Kaplan Meier survival curves were compared between the intervention groups with the use of the log-rank test. Analyses were performed using SAS 9.3 software (SAS Institute, Cary, NC, USA). Results A total of 245 patients were included in the intention to treat analysis (Fig. 1). Patients were randomly allocated to intra-arrest cooling (123 patients) or hospital cooling (122 patients). The baseline characteristics of patients are represented in Table 1. Patients in the IATH group received 1,000 ml (500 1,800) cold saline before hospital admission. When ROSC was achieved, the median (IQR) time to reach a temperature under 34 C (time between ROSC and target temperature achievement) was 22 (10 45) min in the IATH group versus 97 (14 312) min in the hospital cooling group (p = 0.07). The median (IQR) temperature at hospital admission was 33.9 ( ) C in the IATH versus 35 ( ) C in the control group (p = 0.025). The median (IQR) temperature decreased by 1.7 ( ) C between randomization to hospital admission in the IATH versus 0.7 (0 1.5) C for the hospital cooling group (p = 0.015) (Fig. 2a, b). IATH decreased time to reach temperature B34 C by 75 min (95 % CI: 4; 269). Overcooling was frequent during transportation in the IATH group with 17 patients (13.8 %) with at least one a Temperature ( C) b Temperature ( C) temperature recorded under 32 C and 5 patients with a temperature recorded under 30 C. Characteristics and biological data of patients admitted alive to the hospital are presented in Table 2. Outcomes 0 Intra arrest therapeutic hypothermia group group * * * * * Randomization Out of hospital (minutes) Still out of hospital = Deceased = Admitted to hospital = * Intra arrest therapeutic hypothermia group group Hospital admission Hospital (hours)** Still hospitalized = Deceased = Fig. 2 a, b Temperature time course before and after hospital admission (for the 77 patients admitted alive in the hospital) according to group. Time are presented in minutes from randomization for prehospital and in hours after admission for hospital data. Results are presented in median with interquartile range. *p \ 0.05; **Hours reported from hospital admission A total of 77 patients were admitted alive to the hospital: 41 in the IATH group and 36 in the hospital cooling group (p = 0.51). For patients admitted alive, there was no significant difference in the linear mixed effect models analysis for NSE, IL-6, IL-8 and IL-10 (Fig. 3). *

6 Table 2 Characteristics of patient admitted alive to hospital Characteristics Intra-arrest therapeutic hypothermia (41 patients) (36 patients) p Missing values Age, years, median (IQR) 62.4 (51.1; 70.5) 66.8 (49.2; 75.9) Male sex, n (%) 30 (73.2) 20 (55.6) SAPS 2 score 85 (67; 95) 82 (73; 91) Admission characteristics Cathlab admission, n (%) 25 (61) 16 (44.4) Heart rate, median (IQR) 80 (53; 96) 77 (40; 95) Systolic blood pressure, median (IQR) 100 (60; 123) 89.5 (59; 133.5) Diastolic blood pressure, median (IQR) 58 (40; 77) 61.5 (27.5; 75.5) Signs of overload on first chest x-ray, n (%) 7 (17.1) 6 (16.7) Biological data PaO 2 /FiO 2 ratio at admission, median (IQR) (117.5; 330) 140 (86; 266) ph at admission, median (IQR) 7.1 (7; 7.3) 7.1 (7; 7.2) Serum potassium value at admission, mmol/l, median (IQR) 4 (3.4; 5.1) 4.6 (3.7; 5.6) Blood lactate at admission, mmol/l, median (IQR) 7 (3.2; 12.1) 8.2 (3.2; 12.5) Blood glucose value at admission, mmol/l, median (IQR) 12.3 (9.3; 16.7) 13 (10.7; 15.8) Serum creatinine at admission, lmol/l, median (IQR) 120 (98; 145) 127 (112; 146) Serum creatinine at 24 h, lmol/l, median (IQR) 85 (63; 176) 147 (108; 213) Troponin at admission, lg/l, median, (IQR) 0.5 (0.2; 3.6) 0.5 (0.2; 2.5) Troponin at 24 h, lg/l median (IQR) 4.1 (1.4; 26.7) 6.6 (1.2; 38) Creatin kinase at admission, U/L, median (IQR) 354 (184; 513) (142; 807) Creatin kinase at 24 h, U/L, median (IQR) 970 (382; 1,263) 1,984 (235; 3,700) PT at admission, s, median (IQR) 61 (34.5; 71) 49 (24; 66) PT at 24 h, s, median (IQR) 74.5 (62; 81) 67 (41; 76) IQR interquartile range, SAPS simplified acute physiology score, FiO 2 fraction of inspired oxygen, PT prothrombin time The median (IQR) NSE value at 24 h was 96.7 lg/l ( ) in the IATH group versus 97.6 lg/l ( ), p = 0.64 in the hospital cooling group. There were no differences in outcomes globally or regarding the initial rhythm (Table 3; Figs. S1, S2 in the electronic supplement). For patients admitted to the hospital, between the IATH (n = 41) and the hospital cooling group (n = 36), the rate of pulmonary edema (n = 7 vs. 8, p = 0.59), pneumonia (n = 7 vs. 3, p = 0.24), hyperthermia (n = 9 vs. 5, p = 0.36), bacteremia (n = 1 vs. 0, p = 1), hemorrhage (n = 3 vs. 3, p = 0.88), arrhythmia (n = 5 vs. 7, p = 0.39) and convulsion (n = 8 vs. 2, p = 0.06) did not differ during the first 72 h. Discussion In this randomized controlled trial, the induction of IATH decreased the core temperature by an average of 1.7 C prior to hospital admission and shortened the time to reach 34 C by an average of 75 min compared to hospital TH alone. However, this was not associated with differences in markers of neurological injury or in neurological outcome. Targeted temperature management is now an important part of the bundle therapy for the treatment of postcardiac arrest syndrome [27]. Despite a clinical recommendation to induce hypothermia as soon as possible, recent clinical studies have failed to show an improvement in outcomes when hypothermia was induced after ROSC in prehospital setting. In a recent multicenter prospective randomized trial with 1,359 patients, Kim et al. showed that prehospital cooling reduced the time to reach a temperature of 34 C but was not associated with improved survival or neurological status [9]. In two other studies on both shockable or non-shockable rhythms in OHCA, Bernard et al. did not find differences in outcomes when cooling was initiated before hospital arrival [7, 8]. However, time to achieve the targeted temperature in control groups was approximately 60 min and may not reflect clinical practice in most hospitals where time to target temperature has been shown to be longer [5, 28]. Several experimental studies suggested that IATH could be important to limit reperfusion injury [10 13]. A recent clinical study with the use of transnasal evaporative cooling during CPR did not show improvement in outcome except for some subgroup analyses [28]. Our results are consistent with this previous study as IATH was not associated with an improvement of primary and secondary endpoints. NSE is a gamma isomer of enolase that is principally located in neurons and neuroectodermal cells [29]. NSE has been shown to be released in cerebrospinal fluid and in the blood. and to correlate with the extent of brain damage after ischemic, traumatic brain injury, and cardiac arrest [17, 19, 30 35]. However, more recently. the variability among studies in both assays and cutoff values as well as the impact of TH have limited its interest [3, 21,

7 a NSE µg/l NSE Intra arrest therapeutic hypothermia P value = 0.88 Alive= 245 (63) 66 (35) 60 (29) 49 (24) 41 (14) 33 (12) 28 (10) Death= b IL6 ng/l IL6 Intra arrest therapeutic hypothermia P value = 0.74 Alive= 245 (64) 66 (35) 60 (30) 49 (24) 41 (15) 33 (13) 28 (11) Death= Hours Hours c IL8 ng/l IL8 Intra arrest therapeutic hypothermia P value = 0.59 Alive= 245 (64) 66 (35) 60 (30) 49 (24) 41 (15) 33 (13) 28 (11) Death= d IL10 ng/l IL10 Intra arrest therapeutic hypothermia P value = 0.97 Alive= 245 (65) 66 (35) 60 (30) 49 (24) 41 (15) 33 (13) 28 (11) Death= Hours Hours Fig. 3 Evolution of neuron specific enolase (NSE) and interleukins 6, 8 and 10 between IATH and hospital cooling group. n = number of patient at each time point (number of imputed data). Time 0 was randomized 29]. Recent studies support serial measurements of NSE in association with other predictive tools to predict outcome in patients treated with hypothermia [29, 36]. Inflammatory markers such as IL-6, IL-8, and IL-10 have also been shown to increase after cardiac arrest and to correlate to the severity of the post-cardiac syndrome [19]. In our study, in accordance with the absence of clinical benefit, there were no significant differences in value or variation of NSE or IL levels between groups. Cold saline as a cooling method could also have impacted our outcomes. In the Kim et al. prehospital study, cold saline was associated with an increase of rearrest and pulmonary edema [9]. In animal studies, intra-arrest cold saline infusion has also been shown to decrease coronary perfusion pressure compared to external cooling after ROSC [16]. Two multicenter prospective studies are currently recruiting to explore intra-arrest cooling with [37] or without cold saline use (NCT ). Re-arrest rate was not recorded in our study, but the number of patients who achieved ROSC and died before hospital arrival did not differ significantly between groups. Despite a similar decrease in core

8 Table 3 Outcomes data Intra-arrest therapeutic hypothermia (123 patients) (122 patients) p Missing values ROSC n (%) 46 (37.4) 38 (31.1) Admitted alive at hospital n (%) 41 (33.3) 36 (29.5) h survival n (%) 22 (17.9) 19 (15.6) ICU length of stay, days, median (IQR) 3 (1, 8) 2 (1; 5.5) Survival to hospital discharge n (%) 7 (5.7) 5 (4.1) With CPC 1 4 (3.3) 3 (2.5) With CPC 2 3 (2.4) 1 (0.8) With CPC With CPC (0.8) 1-month survival n (%) 7 (5.7) 5 (4.1) With CPC 1 5 (4.1) 3 (2.5) With CPC 2 2 (1.6) 1 (0.8) With CPC With CPC (0.8) 1-year survival n (%) 5 (4.1) 5 (4.1) IQR interquartile range, ROSC return of spontaneous circulation, ICU intensive care unit, CPC cerebral performance category temperature, the volume of cold saline infused was lower in our study (1,000 vs. 2,000 ml in the Kim et al. study [9]); moreover, there was no significant difference in chest x-ray signs of overload or PaO 2 /FiO 2 ratio during the first 24 h between groups. Serum creatinine level at 24 h was significantly improved in the IATH group. Early goal therapy with a multi-system approach is recommended for the treatment of post-cardiac syndrome. Better optimization of volume status in the IATH group could have improved kidney function [3]. Prothrombin time tends to be longer in the IATH group and could reflect some early coagulation disorder in the IATH group [38]. In our study, the hospital targeted temperature in both groups was set between 32 and 34 C for 24 h. The results might have been different if a more conservative approach with normothermia had been targeted in the control group. A recent multicenter prospective study did not shown any differences between a targeted temperature of 33 versus 36 C. Nonetheless, they did not find any harm with a targeted temperature of 33 C as compared to 36 C [4, 39]. Moreover, despite a protocol of continuous temperature monitoring and directives to stop cooling after reaching the target temperature, we observed an increase of overcooled patients in the IATH group. As overcooling has been shown to be deleterious during ICU stay, this could have decreased the potential benefit of the intra-arrest cooling [40]. A higher targeted temperature could have limited overcooling in the IATH group and showed more beneficial outcomes. There are some limitations with this study. First, the inclusion criteria required that patients were still in cardiac arrest at randomization. Patients with ROSC before prehospital EMS arrival, and therefore better prognoses, were excluded from the study. Compared with other prehospital hypothermia trials, patients in our study had longer time to ROSC, lower temperature at hospital admission in the control group, and higher NSE value [7 9, 21]. Patients could therefore have been more severely injured, limiting the potential benefit of the therapy. NSE levels obtained were also higher in mean and in SD than originally hypothesized which led to a decrease in the power of the study. Second, results may have been different with the use of other CPR techniques that have been shown to provide increased circulation and accelerated cooling rate in animals in cardiac arrest, such as active compression decompression CPR with an impedance threshold device [41]. Third, some difficulties in the evaluation of the results of this study may have been induced by the required delay and the variation of the time for initiation of hypothermia in the hospital cooling group. Due to technical difficulties (impossibility to get venous access in the prehospital setting, preservation issues, etc.), some blood samples were missing for some patients. The frequency of missing samples was equally distributed among groups and outcome variables. There were no differences between imputed and raw results. Conclusion Intra-arrest TH by rapid cold saline infusion and external cooling in the prehospital setting decreased temperature by 1.7 C before hospital admission and decreased time to target temperature by 75 min compared to hospital cooling alone. Biological markers of inflammation and brain damage as well as clinical outcomes did not improve with the use of intra-arrest TH. The results of the current study do not support routine use of intra-arrest TH in the prehospital setting to improve clinical outcomes. Results

9 from further studies exploring intra-arrest cooling are warranted. Acknowledgments We thank the emergency physicians, nurses and ambulances crew from Grenoble University Hospital, Annecy Hospital and Voiron Hospital. We also thank the intensive care physicians and nurses of the different intensives care units of those hospitals for their participation in this study. Conflicts of interest Funding for the study was provided by a grant from the French Society of Emergency Medicine (SFMU). The SFMU had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. The authors declare no conflicts of interest. References 1. Deakin CD, Nolan JP, Soar J, Sunde K, Koster RW, Smith GB, Perkins GD (2010) European Resuscitation Council Guidelines for Resuscitation 2010 Section 4. Adult advanced life support. Resuscitation 81: Hypothermia after Cardiac Arrest Study G (2002) Mild therapeutic hypothermia to improve the neurologic outcome after cardiac arrest. N Engl J Med 346: Peberdy MA, Callaway CW, Neumar RW, Geocadin RG, Zimmerman JL, Donnino M, Gabrielli A, Silvers SM, Zaritsky AL, Merchant R, Vanden Hoek TL, Kronick SL, American Heart A (2010) Part 9: post-cardiac arrest care: 2010 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Circulation 122:S768 S Nielsen N, Wetterslev J, Cronberg T, Erlinge D, Gasche Y, Hassager C, Horn J, Hovdenes J, Kjaergaard J, Kuiper M, Pellis T, Stammet P, Wanscher M, Wise MP, Aneman A, Al-Subaie N, Boesgaard S, Bro-Jeppesen J, Brunetti I, Bugge JF, Hingston CD, Juffermans NP, Koopmans M, Kober L, Langorgen J, Lilja G, Moller JE, Rundgren M, Rylander C, Smid O, Werer C, Winkel P, Friberg H, Investigators TTMT (2013) Targeted temperature management at 33 degrees C versus 36 degrees C after cardiac arrest. N Engl J Med 369: Dumas F, Grimaldi D, Zuber B, Fichet J, Charpentier J, Pene F, Vivien B, Varenne O, Carli P, Jouven X, Empana JP, Cariou A (2011) Is hypothermia after cardiac arrest effective in both shockable and nonshockable patients?: insights from a large registry. Circulation 123: Bruel C, Parienti JJ, Marie W, Arrot X, Daubin C, Du Cheyron D, Massetti M, Charbonneau P (2008) Mild hypothermia during advanced life support: a preliminary study in out-ofhospital cardiac arrest. Crit Care 12:R31 7. Bernard SA, Smith K, Cameron P, Masci K, Taylor DM, Cooper DJ, Kelly AM, Silvester W, Rapid Infusion of Cold Hartmanns I (2010) Induction of therapeutic hypothermia by paramedics after resuscitation from out-of-hospital ventricular fibrillation cardiac arrest: a randomized controlled trial. Circulation 122: Bernard SA, Smith K, Cameron P, Masci K, Taylor DM, Cooper DJ, Kelly AM, Silvester W, Rapid Infusion of Cold Hartmanns I (2012) Induction of prehospital therapeutic hypothermia after resuscitation from nonventricular fibrillation cardiac arrest. Crit Care Med 40: Kim F, Nichol G, Maynard C, Hallstrom A, Kudenchuk PJ, Rea T, Copass MK, Carlbom D, Deem S, Longstreth WT Jr, Olsufka M, Cobb LA (2014) Effect of prehospital induction of mild hypothermia on survival and neurological status among adults with cardiac arrest: a randomized clinical trial. J Am Med Assoc 311: Abella BS, Zhao D, Alvarado J, Hamann K, Vanden Hoek TL, Becker LB (2004) Intra-arrest cooling improves outcomes in a murine cardiac arrest model. Circulation 109: Nozari A, Safar P, Stezoski SW, Wu X, Kostelnik S, Radovsky A, Tisherman S, Kochanek PM (2006) Critical time window for intra-arrest cooling with cold saline flush in a dog model of cardiopulmonary resuscitation. Circulation 113: Zhao D, Abella BS, Beiser DG, Alvarado JP, Wang H, Hamann KJ, Hoek TL, Becker LB (2008) Intra-arrest cooling with delayed reperfusion yields higher survival than earlier normothermic resuscitation in a mouse model of cardiac arrest. Resuscitation 77: Wang H, Barbut D, Tsai MS, Sun S, Weil MH, Tang W (2010) Intra-arrest selective brain cooling improves success of resuscitation in a porcine model of prolonged cardiac arrest. Resuscitation 81: Darbera L, Chenoune M, Lidouren F, Kohlhauer M, Adam C, Bruneval P, Ghaleh B, Dubois-Randé J-L, Carli P, Vivien B (2013) Hypothermic liquid ventilation prevents early hemodynamic dysfunction and cardiovascular mortality after coronary artery occlusion complicated by cardiac arrest in rabbits. Crit Care Med 41:e457 e Menegazzi JJ, Rittenberger JC, Suffoletto BP, Logue ES, Salcido DD, Reynolds JC, Sherman LD (2009) Effects of pre-arrest and intra-arrest hypothermia on ventricular fibrillation and resuscitation. Resuscitation 80: Yannopoulos D, Zviman M, Castro V, Kolandaivelu A, Ranjan R, Wilson RF, Halperin HR (2009) Intracardiopulmonary resuscitation hypothermia with and without volume loading in an ischemic model of cardiac arrest. Circulation 120: Wijdicks EF, Hijdra A, Young GB, Bassetti CL, Wiebe S, Quality Standards Subcommittee of the American Academy of N (2006) Practice parameter: prediction of outcome in comatose survivors after cardiopulmonary resuscitation (an evidence-based review): report of the Quality Standards Subcommittee of the American Academy of Neurology. Neurology 67:

10 18. Neumar RW, Nolan JP, Adrie C, Aibiki M, Berg RA, Bottiger BW, Callaway C, Clark RS, Geocadin RG, Jauch EC, Kern KB, Laurent I, Longstreth WT Jr, Merchant RM, Morley P, Morrison LJ, Nadkarni V, Peberdy MA, Rivers EP, Rodriguez-Nunez A, Sellke FW, Spaulding C, Sunde K, Vanden Hoek T (2008) Post-cardiac arrest syndrome: epidemiology, pathophysiology, treatment, and prognostication. A consensus statement from the International Liaison Committee on Resuscitation (American Heart Association, Australian and New Zealand Council on Resuscitation, European Resuscitation Council, Heart and Stroke Foundation of Canada, InterAmerican Heart Foundation, Resuscitation Council of Asia, and the Resuscitation Council of Southern Africa); the American Heart Association Emergency Cardiovascular Care Committee; the Council on Cardiovascular Surgery and Anesthesia; the Council on Cardiopulmonary, Perioperative, and Critical Care; the Council on Clinical Cardiology; and the Stroke Council. Circulation 118: Adrie C, Adib-Conquy M, Laurent I, Monchi M, Vinsonneau C, Fitting C, Fraisse F, Dinh-Xuan AT, Carli P, Spaulding C, Dhainaut JF, Cavaillon JM (2002) Successful cardiopulmonary resuscitation after cardiac arrest as a sepsis-like syndrome. Circulation 106: Calderon LM, Guyette FX, Doshi AA, Callaway CW, Rittenberger JC, Post Cardiac Arrest S (2014) Combining NSE and S100B with clinical examination findings to predict survival after resuscitation from cardiac arrest. Resuscitation 85(8): Oksanen T, Tiainen M, Skrifvars MB, Varpula T, Kuitunen A, Castren M, Pettila V (2009) Predictive power of serum NSE and OHCA score regarding 6-month neurologic outcome after outof-hospital ventricular fibrillation and therapeutic hypothermia. Resuscitation 80: Tiainen M, Roine RO, Pettila V, Takkunen O (2003) Serum neuronspecific enolase and S-100B protein in cardiac arrest patients treated with hypothermia. Stroke 34: Recommended guidelines for uniform reporting of data from out-ofhospital cardiac arrest: the Utstein style. Prepared by a Task Force of Representatives from the European Resuscitation Council, American Heart Association, Heart and Stroke Foundation of Canada, Australian Resuscitation Council. Resuscitation 22: Gueugniaud PY, David JS, Chanzy E, Hubert H, Dubien PY, Mauriaucourt P, Braganca C, Billeres X, Clotteau- Lambert MP, Fuster P, Thiercelin D, Debaty G, Ricard-Hibon A, Roux P, Espesson C, Querellou E, Ducros L, Ecollan P, Halbout L, Savary D, Guillaumee F, Maupoint R, Capelle P, Bracq C, Dreyfus P, Nouguier P, Gache A, Meurisse C, Boulanger B, Lae C, Metzger J, Raphael V, Beruben A, Wenzel V, Guinhouya C, Vilhelm C, Marret E (2008) Vasopressin and epinephrine vs. epinephrine alone in cardiopulmonary resuscitation. N Engl J Med 359: Vesin A, Azoulay E, Ruckly S, Vignoud L, Rusinovà K, Benoit D, Soares M, Azeivedo-Maia P, Abroug F, Benbenishty J, Timsit JF (2013) Reporting and handling missing values in clinical studies in intensive care units. Intensive Care Med 39: Raghunathan TE, Lepkowski JM, Van Hoewyk J, Solenberger P (2001) A multivariate technique for multiply imputing missing values using a sequence of regression models. Survey Methodol 27: Neumar RW, Otto CW, Link MS, Kronick SL, Shuster M, Callaway CW, Kudenchuk PJ, Ornato JP, McNally B, Silvers SM, Passman RS, White RD, Hess EP, Tang W, Davis D, Sinz E, Morrison LJ (2010) Part 8: adult advanced cardiovascular life support: 2010 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Circulation 122:S729 S Castren M, Nordberg P, Svensson L, Taccone F, Vincent JL, Desruelles D, Eichwede F, Mols P, Schwab T, Vergnion M, Storm C, Pesenti A, Pachl J, Guerisse F, Elste T, Roessler M, Fritz H, Durnez P, Busch HJ, Inderbitzen B, Barbut D (2010) Intra-arrest transnasal evaporative cooling: a randomized, prehospital, multicenter study (PRINCE: Pre-ROSC IntraNasal Cooling Effectiveness). Circulation 122: Rundgren M, Karlsson T, Nielsen N, Cronberg T, Johnsson P, Friberg H (2009) Neuron specific enolase and S-100B as predictors of outcome after cardiac arrest and induced hypothermia. Resuscitation 80: Cunningham RT, Young IS, Winder J, O Kane MJ, McKinstry S, Johnston CF, Dolan OM, Hawkins SA, Buchanan KD (1991) Serum neurone specific enolase (NSE) levels as an indicator of neuronal damage in patients with cerebral infarction. Eur J Clin Invest 21: Skogseid IM, Nordby HK, Urdal P, Paus E, Lilleaas F (1992) Increased serum creatine kinase BB and neuron specific enolase following head injury indicates brain damage. Acta Neurochir (Wien) 115: Martens P, Raabe A, Johnsson P (1998) Serum S-100 and neuron-specific enolase for prediction of regaining consciousness after global cerebral ischemia. Stroke 29: Meynaar IA, Oudemans-van Straaten HM, van der Wetering J, Verlooy P, Slaats EH, Bosman RJ, van der Spoel JI, Zandstra DF (2003) Serum neuronspecific enolase predicts outcome in post-anoxic coma: a prospective cohort study. Intensive Care Med 29: Rosen H, Sunnerhagen KS, Herlitz J, Blomstrand C, Rosengren L (2001) Serum levels of the brain-derived proteins S-100 and NSE predict longterm outcome after cardiac arrest. Resuscitation 49: Schoerkhuber W, Kittler H, Sterz F, Behringer W, Holzer M, Frossard M, Spitzauer S, Laggner AN (1999) Time course of serum neuron-specific enolase. A predictor of neurological outcome in patients resuscitated from cardiac arrest. Stroke 30: Huntgeburth M, Adler C, Rosenkranz S, Zobel C, Haupt WF, Dohmen C, Reuter H (2014) Changes in neuron-specific enolase are more suitable than its absolute serum levels for the prediction of neurologic outcome in hypothermiatreated patients with out-of-hospital cardiac arrest. Neurocrit Care 20: Deasy C, Bernard S, Cameron P (2011) Design of the RINSE trial: the rapid infusion of cold normal saline by paramedics during CPR. BMC Emerg Med 11: Polderman KH, Herold I (2009) Therapeutic hypothermia and controlled normothermia in the intensive care unit: practical considerations, side effects, and cooling methods. Crit Care Med 37:

11 39. Annborn M, Bro-Jeppesen J, Nielsen N, Ullén S, Kjaergaard J, Hassager C, Wanscher M, Hovdenes J, Pellis T, Pelosi P, Wise MP, Cronberg T, Erlinge D, Friberg H (2014) The association of targeted temperature management at 33 and 36 C with outcome in patients with moderate shock on admission after out-of-hospital cardiac arrest: a post hoc analysis of the target temperature management trial. Intensive Care Med 40: Merchant RM, Abella BS, Peberdy MA, Soar J, Ong ME, Schmidt GA, Becker LB, Vanden Hoek TL (2006) Therapeutic hypothermia after cardiac arrest: unintentional overcooling is common using ice packs and conventional cooling blankets. Crit Care Med 34:S490 S Srinivasan V, Nadkarni VM, Yannopoulos D, Marino BS, Sigurdsson G, McKnite SH, Zook M, Benditt DG, Lurie KG (2006) Rapid induction of cerebral hypothermia is enhanced with active compression-decompression plus inspiratory impedance threshold device cardiopulmonary resusitation in a porcine model of cardiac arrest. J Am Coll Cardiol 47: