Remember me
This Invited Commentary accompanies the following article: Behringer W, Böttiger BW, Biasucci DG et al. Temperature control after successful resuscitation from cardiac arrest in adults: a joint statement from the European Society for Emergency Medicine and the European Society of Anaesthesiology and Intensive Care. Eur J Anaesthesiol. 2024; 41:278-281.
We read with some surprise the recent joint statement from the European Society for Emergency Medicine (EUSEM) and the European Society of Anaesthesiology and Intensive Care (ESAIC) on temperature control after successful resuscitation from cardiac arrest in adults.1,2 In this statement, the authors recommend that ‘… clinicians should consider hypothermia in the range of 32 to 34°C in all adults after cardiac arrest as soon as feasible’.
This recommendation conflicts with the 2022 Consensus on Science With Treatment Recommendations from the International Liaison Committee on Resuscitation (ILCOR)3 and joint guidelines from the European Resuscitation Council (ERC) and the European Society of Intensive Care Medicine (ESICM)4 which suggests ‘… actively preventing fever by targeting a temperature of 37.5 °C for patients who remain comatose after return of spontaneous circulation from cardiac arrest’. These recommendations were based on a rigorous methodological process including systematic reviews with meta-analyses and use of the GRADE methodology to evaluate certainty in the evidence and to translate the evidence into guidelines.5 The original ILCOR systematic review performed by us was published in 2021 and updated in 2023.6–8 The methodology used to inform the EUSEM and ESAIC recommendations is not described, including how potential or actual author conflicts of interests were managed.1,2
Why are the EUSEM/ESAIC recommendations different from those of ILCOR, ERC and ESICM? The authors provide three arguments: a recent Cochrane review found a statistically significant benefit for neurological outcome of therapeutic hypothermia in the range of 32 to 34 °C,9 which was contrary to the ILCOR systematic reviews6–8; the two large ‘Targeted temperature management (TTM) trials’10,11 have certain limitations that make them less generalisable than other trials, and animal studies and some observational studies support a beneficial effect of hypothermia. We will address each of these below.
We have previously described the methodological differences between the Cochrane review and our updated ILCOR review.7 The primary difference is the inclusion of four small older trials in the Cochrane review. These trials had either a very short duration of temperature control (<4 h because rewarming was started as soon as the target temperature was reached),12,13 a very long duration of temperature management (72 h),14 or were published only as an abstract.15 In our opinion, the inclusion of these small trials with lower methodological quality and increased heterogeneity does not increase the validity or generalisability of the results from the meta-analysis. Even with the inclusion of these trials, the results from the fixed effects meta-analyses in the Cochrane and ILCOR reviews are largely comparable. In the ILCOR review, the risk ratio from the meta-analysis of favourable neurological outcome at any time point was 1.14 [95% confidence interval (CI), 0.98 to 1.34].7,8 In the Cochrane review, the risk ratio was 1.09 (95% CI, 1.02 to 1.18).9 The conclusion from these analyses should not be substantially different. They both show a potential small effect of hypothermia but with uncertainty in the effect estimate. An uncertainty that includes no clinically meaningful effect in both analyses. The primary results published in the Cochrane review came from a random effects meta-analysis, which puts relatively more weight on the small older trials with extreme results.
Guidelines are based on many considerations other than the results of a meta-analysis, including an assessment of the internal validity and generalisability of the individual trials. Most meta-analyses simply pool trials based on the size of the trial (or the precision of the results) and ignore the individual quality of the trials. We believe that the larger more recent trials [i.e. Nielsen et al. (TTM-1), Lascarrou et al., Dankiewicz et al. (TTM-2), and Wolfrum et al.]19 are of a higher methodological quality, and better reflect contemporary postcardiac arrest care, compared with earlier trials,10,16,17 many of which included patients more than 20 years ago.18
The EUSEM/ESAIC recommendations list three main limitations of the two large TTM trials: high bystander cardiopulmonary resuscitation (CPR) rates and shorter no-flow time; delay in starting temperature control and reaching target temperature; and inclusion of many centres from multiple countries potentially resulting in heterogeneity.
The authors note that the bystander CPR rate in the two TTM trials was 73 to 82% and quote an average rate in Europe of 58%. The rate of bystander CPR of 58% is based on the EuReCa TWO study.19 However, this study uses a completely different denominator, namely all patients with out-of-hospital cardiac arrest. The TTM trials included only those with cardiac arrest of a presumed cardiac origin and with return of spontaneous circulation (ROSC) who survived to the intensive care unit (ICU) and were randomised. As bystander CPR is associated with more favourable outcomes, it is not surprising that those who survive to ICU admission and randomisation have a higher proportion of bystander CPR. Moreover, the TTM investigators have conducted several subgroup analyses to examine potential effect measure heterogeneity, including according to bystander CPR, no-flow times, low-flow times, and other measures of severity.11,20,21 For most of these analyses, there has been no sign of effect measure heterogeneity. For bystander CPR, a combined analysis of the two TTM trials found potential effect measure heterogeneity with improved outcomes with hypothermia in those without bystander CPR and improved outcomes with normothermia in those with bystander CPR.21 However, the results were uncertain (i.e. wide confidence intervals) and such post hoc subgroup analyses should be interpreted carefully. Either way, these subgroups do not support a general recommendation favouring hypothermia as most patients eligible for temperature control in Europe nowadays receive bystander CPR. Furthermore, we note that a recent individual participant data meta-analysis of TTM-2 and the Lascarrou et al. trial found that hypothermia at 33 °C did not significantly improve survival or functional outcome in patient with an initial nonshockable rhythm (another risk factor for poor outcomes).22
The authors also note that there was a delay in the initiation of temperature control and a delay in reaching the target temperature in the two TTM trials. We believe this reflects current clinical practice. Table 1 provides an overview of the approximate time from ROSC to randomisation, the time from randomisation to temperature target, and the time from ROSC to temperature target from all the trials included in the ILCOR meta-analysis. Not all times were reported in the manuscripts, and some had to be estimated from figures, but the picture is quite clear. All the trials (except the small Bernard et al.23 trial where cooling was started prehospital) had a delay. This is not only the case in these specific clinical trials. In multicentre, observational studies, large average delays (i.e., most often 5 to 6 h from ROSC to temperature target) have also been reported.29–35 Similar delays in reaching target temperature are seen in clinical trials testing other aspects of postcardiac arrest care.36–39 As such, this is not a weakness of the two TTM trials, but simply a reflection of clinical practice. Although it is possible that very early cooling might be beneficial, we note that trials of prehospital cooling have not shown a benefit.6 An exploratory analysis from the TTM-2 investigators found no benefit of a lower temperature when examining sites with the fastest cooling times in the TTM-2 trial.40 This explanation is, therefore, unlikely to explain the neutral results seen in the TTM trials.
Table 1 - Trials assessing TTM at 32 to 34 °C included in the recent ILCOR meta-analysis Trial (first author, year) Target Time from ROSC to randomisation Time from randomisation to target Time from ROSC to target Bernard, 200223 33 °C NR NR 2 h HACA, 200224 32 to 34 °C 105 mina NR 8 h Laurent, 200525 32 °C NR NR NR Hachimi-Idrissi, 200513 33 °C NR NR NR Nielsen, 201310 33 °C NR 3 h to <34 °Cb NR Lascarrou, 201926 33 °C 233 min 317 min to 33 °C 9 h Dankiewicz, 202127 33 °C 111 min 3 h to 34 °C 5 h Kwon, 202128 33 °C NR NR 5 to 6 hb Wolfrum, 202218 33 °C 2 ha 4.2 h to <34 °C 6 hILCOR, International Liason Comittee on Resuscitation; NR, not reported; ROSC, return of spontaneous circulation; TTM, Targeted temperature management.
aTime to initiation of cooling from ROSC.
bEstimated from figure.
We consider the inclusion of many centres from multiple countries to be a strength of the TTM trials, not a weakness.
The critique of the TTM trials is peculiar as there is no similar critique of the other trials included in the Cochrane meta-analysis. The main trials that showed benefit of hypothermia were the HACA-trial and the trial by Bernard et al.,23,24 both published in 2002. In Table 2, we provide a comparison of these trials and the TTM-2 trial.27 Although the 2002 trials were ground-breaking when published, they do not have the same methodological rigour as many of the later trials, including the TTM-2 trial. Both internal validity and potential for generalisability are clearly much higher for the TTM-2 trial. Arguing that patients in the TTM-2 trial were not as sick as those in the early trials or had less brain injury is unwarranted because many of the cardiac characteristics other than bystander CPR (e.g. initial rhythm, witnessed status, time to ROSC) were similar or less favourable in the TTM-2 trial. Moreover, despite a general improvement in outcomes for cardiac arrest patients over time, outcomes in the control groups are similar in the trials, indicating that patients included in the TTM-2 trial were as sick as those included in the earlier trials.
Table 2 - Comparison of the trial by Bernard et al., the HACA-trial and the TTM-2-trial Bernard et al.23 HACA24 TTM-227 Description Sites 1 centre in Australia 9 centres from 5 countries in Europe >50 centres from 14 countries primarily in Europe and Australia Years of enrolment 1996 to 1999 1996 to 2001 2017 to 2020 Patients screened NR 3551 4355 Patients included 77a 275b 1861 Cardiac arrest characteristics Shockable rhythm 100% 100% 74% Witnessed 95% 98% 91% Bystander CPR 58% 46% 80% Time to ROSC Mean: 26 min Median: 21 to 22 min Median: 25 min Intervention 33°C for 12 h, started in the ambulance 32 to 34 °C for 24 h 33 °C for 28 h after randomization Control 37 °C Normothermiac ≤37.5 °C Registration and bias Preregistration and protocol No No Yes Randomisation Pseudo-randomd Yes Yes Allocation concealment No Yes Yes Blinding No No, but blinded assessment of neurological outcomes No, but blinded assessment of neurologic prognosis and neurological outcomes Outcomes Primary outcome Survival to hospital discharge home or to rehabilitation Favourable neurological outcome within 6 months Death from any cause at 6 months Primary result OReData are presented as n, %, and OR (95% CI), RR (95% CI). CI, confidence interval; CPR: Cardiopulmonary resucitation; ROSC, return of spontaneous circulation; OR, odds ratio; RR, risk ratio.
aThe original sample size was 62 patients. After an analysis of the 62 patients, which showed a ‘… strong trend toward improved outcome in the hypothermia group …’, the trial was continued.
bStopped early because of low enrolment rates. Original sample size not reported.
cThe precise description in the manuscript is ‘Patients randomly assigned to the normothermia group were placed on a conventional hospital bed, and normothermia was maintained’.
dPatients were assigned based on odd and even days.
eAdjusted for age and time to ROSC. The P value is unadjusted.
The authors of the EUSEM and ESAIC recommendations cite several animal models and observational studies to support their recommendations. Although we believe that animal models play an important role in the advancement of cardiac arrest science,41 we also believe that they should not be used to inform clinical guidelines unless there are no human data. This is not the case for temperature control after cardiac arrest. Animal studies often suffer from poor reporting and a high risk of bias42 and rarely reflect the human clinical condition.43 Even if animal studies have internal validity, it is unlikely that results can be easily extrapolated to humans.
Similarly, observational studies play an important role in advancing cardiac arrest science but have an inherent risk of bias primarily because of confounding. This is especially true for a complex intervention such as temperature control. Assuming no unmeasured or residual confounding is a very strong assumption that we consider unlikely to be true in this context. Again, observational studies should be used only to support clinical guidelines if no, or few interventional trials are available. This is not the case for temperature management after cardiac arrest.
Lastly, the citation of animal studies and observational studies appears to be somewhat arbitrary. As far as we can tell, the authors have not performed any updated systematic reviews or meta-analyses with bias assessment of these studies (the Cochrane review only included interventional trials9). This risks selective reporting of specific studies supporting the recommendation.
Clinical guidelines should be evidence-based and consider the values and preferences of stakeholders. When there is a choice of options with no clear superior strategy, the easiest to implement and least interventional strategy is usually preferred. Evidence is best evaluated through a rigorous process, involving systematic reviews including bias assessment and meta-analyses, and a systematic evaluation of the uncertainty in the available evidence, for example, using the GRADE methodology. As the evaluation of the evidence is partly subjective, such evaluation is best done by a broad group of experts and stakeholders, including managing potential conflicts of interest.
Based on our systematic reviews.6–8 and a critical evaluation of the available evidence, we consider the large clinical trials published within the last 10 years10,18,26,27 to be the most valid and with the highest potential for generalisability to current cardiac arrest patients. We do not agree with the recent guidelines by EUSEM and ESAIC, which are based on smaller and older trials as well as animal models and observational studies. Instead, we encourage clinicians to follow the guidelines provided by ERC and ESICM and based on a consensus on science from ILCOR.3,4 Like the authors of the EUSEM and ESAIC guidelines, we encourage additional research on this important topic.
Acknowledgements relating to this articleAssistance with the article: none.
Financial support and sponsorship: none.
Presentation: none.
Conflicts of interest: JN: Editor-in-Chief Resuscitation (honorarium paid by Elsevier), past co-chair ILCOR. JS: Editor Resuscitation, receive payment from publisher Elsevier. Past chair ILCOR ALS Task Force, current co-chair of ERC ALS Science and Education Committee. AG: member of data and safety monitoring board and received personal fees from Noorik Biopharmaceuticals (2021 to 2022). Paid consultant NMD Pharma.
This manuscript was handled by Nicolas Bruder.
Comment from the Editor: this article was checked and accepted by the Editors, but was not sent for external peer-review.
References 1. Behringer W, Bottiger BW, Biasucci DG, et al. Temperature control after successful resuscitation from cardiac arrest in adults: a joint statement from the European Society for Emergency Medicine and the European Society of Anaesthesiology and Intensive Care. Eur J Anaesthesiol 2023; 41:278–281. 2. Behringer W, Bottiger BW, Biasucci DG, et al. Temperature control after successful resuscitation from cardiac arrest in adults: a joint statement from the European Society for Emergency Medicine (EUSEM) and the European Society of Anaesthesiology and Intensive Care (ESAIC). Eur J Emerg Med 2023; 31:86–89. 3. Wyckoff MH, Greif R, Morley PT, et al. Collaborators. 2022 International Consensus on Cardiopulmonary Resuscitation and Emergency Cardiovascular Care Science With Treatment Recommendations: summary from the basic life support; advanced life support; pediatric life support; neonatal life support; education, implementation, and teams; and first aid task forces. Resuscitation 2022; 181:208–288. 4. Nolan JP, Sandroni C, Andersen LW, et al. ERC-ESICM guidelines on temperature control after cardiac arrest in adults. Resuscitation 2022; 172:229–236. 5. Guyatt GH, Oxman AD, Vist GE, et al. GRADE: an emerging consensus on rating quality of evidence and strength of recommendations. BMJ 2008; 336:924–926. 6. Granfeldt A, Holmberg MJ, Nolan JP, et al. International Liaison Committee on Resuscitation Advanced Life Support Task Force. Targeted temperature management in adult cardiac arrest: systematic review and meta-analysis. Resuscitation 2021; 167:160–172. 7. Granfeldt A, Holmberg MJ, Nolan JP, et al. International Liaison Committee on Resuscitation IALSTF. Temperature control after adult cardiac arrest: an updated systematic review and meta-analysis. Resuscitation 2023; 191:109928. 8. Granfeldt A, Holmberg MJ, Nolan JP, et al. International Liaison Committee on Resuscitation IALSTF. Corrigendum to “Temperature control after adult cardiac arrest: an updated systematic review and meta-analysis” [Resuscitation 191 (2023)109928]. Resuscitation 2023; 194:110052. 9. Arrich J, Schutz N, Oppenauer J, et al. Hypothermia for neuroprotection in adults after cardiac arrest. Cochrane Database Syst Rev 2023; 5:CD004128. 10. Nielsen N, Wetterslev J, Cronberg T, et al. Targeted temperature management at 33 degrees C versus 36 degrees C after cardiac arrest. N Engl J Med 2013; 369:2197–2206. 11. Dankiewicz J, Cronberg T, Lilja G, et al. Targeted hypothermia versus targeted Normothermia after out-of-hospital cardiac arrest (TTM2): a randomized clinical trial-Rationale and design. Am Heart J 2019; 217:23–31. 12. Hachimi-Idrissi S, Corne L, Ebinger G, et al. Mild hypothermia induced by a helmet device: a clinical feasibility study. Resuscitation 2001; 51:275–281. 13. Hachimi-Idrissi S, Zizi M, Nguyen DN, et al. The evolution of serum astroglial S-100 beta protein in patients with cardiac arrest treated with mild hypothermia. Resuscitation 2005; 64:187–192. 14. Zhang J. Comparison and evaluation of brain-functional protection with mild hypothermia technique and normal temperature technique. Chin J Clin Rehabil 2005; 92:136–138. 15. Mori K, Takeyama Y, Itoh Y, et al. A multivariate analysis of prognostic factors in survivors of out-of-hospital cardiac arrest with brain hypothermia. Crit Care Med 2000; 28:A168. 16. Lascarrou J-B, Merdji H, Le Gouge A, et al. Targeted temperature management for cardiac arrest with nonshockable rhythm. New Engl J Med 2019; 381:2327–2337. 17. Dankiewicz J, Cronberg T, Lilja G, et al. Hypothermia versus normothermia after out-of-hospital cardiac arrest. New Engl J Med 2021; 384:2283–2294. 18. Wolfrum S, Roedl K, Hanebutte A, et al. Temperature control after in-hospital cardiac arrest: a randomized clinical trial. Circulation 2022; 146:1357–1366. 19. Grasner JT, Wnent J, Herlitz J, et al. Survival after out-of-hospital cardiac arrest in Europe - results of the EuReCa TWO study. Resuscitation 2020; 148:218–226. 20. Dankiewicz J, Friberg H, Belohlavek J, et al. Time to start of cardiopulmonary resuscitation and the effect of target temperature management at 33 degrees C and 36 degrees C. Resuscitation 2016; 99:44–49. 21. Holgersson J, Meyer MAS, Dankiewicz J, et al. Hypothermic versus normothermic temperature control after cardiac arrest. NEJM Evid 2022; 1. 22. Taccone FS, Dankiewicz J, Cariou A, et al. Hypothermia vs normothermia in patients with cardiac arrest and nonshockable rhythm: a meta-analysis. JAMA Neurol 2023; 81:126–133. 23. Bernard SA, Gray TW, Buist MD, et al. Treatment of comatose survivors of out-of-hospital cardiac arrest with induced hypothermia. N Engl J Med 2002; 346:557–563. 24. Hypothermia after Cardiac Arrest Study Group. Mild therapeutic hypothermia to improve the neurologic outcome after cardiac arrest. N Engl J Med 2002; 346:549–556. 25. Laurent I, Adrie C, Vinsonneau C, et al. High-volume hemofiltration after out-of-hospital cardiac arrest: a randomized study. J Am Coll Cardiol 2005; 46:432–437. 26. Lascarrou JB, Merdji H, Le Gouge A, et al. Targeted temperature management for cardiac arrest with nonshockable rhythm. N Engl J Med 2019; 381:2327–2337. 27. Dankiewicz J, Cronberg T, Lilja G, et al. Hypothermia versus normothermia after out-of-hospital cardiac arrest. N Engl J Med 2021; 384:2283–2294. 28. Kwon WY, Jung YS, Suh GJ, et al. Regional cerebral oxygen saturation in cardiac arrest survivors undergoing targeted temperature management 36 degrees C versus 33 degrees C: A randomized clinical trial. Resuscitation 2021; 167:362–371. 29. Nielsen N, Hovdenes J, Nilsson F, et al. Hypothermia Network. Outcome, timing and adverse events in therapeutic hypothermia after out-of-hospital cardiac arrest. Acta Anaesthesiol Scand 2009; 53:926–934. 30. Perman SM, Ellenberg JH, Grossestreuer AV, et al. Shorter time to target temperature is associated with poor neurologic outcome in postarrest patients treated with targeted temperature management. Resuscitation 2015; 88:114–119. 31. Khera R, Humbert A, Leroux B, et al. Hospital variation in the utilization and implementation of targeted temperature management in out-of-hospital cardiac arrest. Circ Cardiovasc Qual Outcomes 2018; 11:e004829. 32. Sonder P, Janssens GN, Beishuizen A, et al. Efficacy of different cooling technologies for therapeutic temperature management: a prospective intervention study. Resuscitation 2018; 124:14–20. 33. Sawyer KN, Mooney M, Norris G, et al. COOL-ARREST: results from a pilot multicenter, prospective, single-arm observational trial to assess intravascular temperature management in the treatment of cardiac arrest. Ther Hypothermia Temp Manag 2019; 9:56–62. 34. Okazaki T, Hifumi T, Kawakita K, et al. Japanese Association for Acute Medicine out-of-hospital cardiac arrest registry. Targeted temperature management guided by the severity of hyperlactatemia for out-of-hospital cardiac arrest patients: a post hoc analysis of a nationwide, multicenter prospective registry. Ann Intensive Care 2019; 9:127. 35. Hifumi T, Inoue A, Kokubu N, et al. Association between rewarming duration and neurological outcome in out-of-hospital cardiac arrest patients receiving therapeutic hypothermia. Resuscitation 2020; 146:170–177. 36. Deye N, Cariou A, Girardie P, et al. Endovascular versus external targeted temperature management for patients with out-of-hospital cardiac arrest: a randomized, controlled study. Circulation 2015; 132:182–193. 37. Kirkegaard H, Soreide E, de Haas I, et al. Targeted temperature management for 48 vs 24 hours and neurologic outcome after out-of-hospital cardiac arrest: a randomized clinical trial. JAMA 2017; 318:341–350. 38. Lemkes JS, Janssens GN, van der Hoeven NW, et al. Coronary angiography after cardiac arrest without ST-segment elevation. N Engl J Med 2019; 380:1397–1407. 39. Francois B, Cariou A, Clere-Jehl R, et al. Prevention of early ventilator-associated pneumonia after cardiac arrest. N Engl J Med 2019; 381:1831–1842. 40. Simpson RFG, Dankiewicz J, Karamasis GV, et al. Speed of cooling after cardiac arrest in relation to the intervention effect: a sub-study from the TTM2-trial. Crit Care 2022; 26:356. 41. Andersen LW, Vammen L, Granfeldt A. Animal research in cardiac arrest. Resusc Plus 2024; 17:100511. 42. Fabian-Jessing BK, Vallentin MF, Secher N, et al. Animal models of cardiac arrest: a systematic review of bias and reporting. Resuscitation 2018; 125:16–21. 43. Vognsen M, Fabian-Jessing BK, Secher N, et al. Contemporary animal models of cardiac arrest: a systematic review. Resuscitation 2017; 113:115–123.
Comments (0)