Mild therapeutic hypothermia has been shown to be beneficial for neurologic outcome and survival for patients after successful resuscitation from sudden cardiac arrest (Arrich 2012), but it remains unclear when this intervention should be initiated.
Description of the condition
The incidence of out-of-hospital sudden cardiac arrest is not easily determined. A review of emergency medicine services (EMS)-treated out-of-hospital cardiac arrest in Europe speaks of an incidence of 41 per 100,000 person-years (Berdowski 2010), while in North America around 52 per 100,000 person-years are reported (Nichol 2008). The survival rates of 9% to 11% for all-rhythm cardiac arrests and 19% to 21% for ventricular fibrillation cardiac arrests seem to have improved in the past decades, but are still disastrously low. When the circulation stops the brain cell is directly damaged through the lack of oxygen and adenosine triphosphate (ATP), but even after the circulation is re-established continuous damage of the brain cell happens through neuronal necrosis and apoptosis caused by a myriad of pathophysiological mechanisms, microcirculatory failure, and other factors such as pyrexia, hyperglycaemia, and seizures (Nolan 2008; Holzer 2010). After resuscitation and admission at a hospital, post-resuscitation care is aimed at reducing the secondary reperfusion injuries caused by the cardiac arrest. This treatment includes early treatment of the cause of the arrest (the majority have a coronary heart disease, thus coronary angiography with subsequent percutaneous coronary intervention (PCI) if indicated), therapeutic hypothermia, optimisation of oxygenation, ventilation, circulation and metabolism, and early seizure detection and treatment following a standardised treatment protocol (Deakin 2010; Peberdy 2010).
Description of the intervention
Therapeutic hypothermia may improve survival and reduce the amount of neurologic damage after cardiac arrest. According to the recent guidelines, comatose survivors of out-of-hospital ventricular fibrillation cardiac arrest should be cooled with internal or external cooling techniques to a target temperature of 32 °C to 34 °C (patients with in-hospital cardiac arrest or other primary rhythms may also benefit) (Deakin 2010; Peberdy 2010). This target temperature should be maintained for 12 to 24 hours, and after this cooling period the patients should be rewarmed at a rate of 0.25 °C to 0.5 °C/hour to normothermia). Early (prehospital) application of mild therapeutic hypothermia refers to the reduction of body core temperature to 32 °C to 34 °C for at least 12 hours by the healthcare professionals in the field, either during resuscitation or shortly after resuscitation. Several methods are available to decrease body temperature, including surface cooling methods such as ice packs, cold-air mattresses, water-circulating cooling pads and pre-cooled cooling pads. The core temperature may also be decreased by intravascular cooling catheters, or by the infusion of cold fluids (Holzer 2010). Other methods include cooling caps or the application of coolant through the nose, which cools the nasal cavity and subsequently the whole body through evaporation (Castren 2010). However, not all methods that are available after admission to a hospital will be available or practical in the field. Cooling should be continued after hospital admission. Maintenance cooling and rewarming should be according to the current guidelines on resuscitation at that time and comparable to the control group.
Formally the intervention in this review is the time point of initiation of mild therapeutic hypothermia. Practically, initiation of therapeutic hypothermia includes significant variation and we expect study protocols and clinical routine to differ substantially. Therefore we will pragmatically compare early initiation in the prehospital phase to later initiation after hospital admission.
How the intervention might work
Pathophysiologically, brain damage through hypoxia is due to two main mechanisms. The first is direct excitotoxic and ischaemic cell death leading to necrosis and apoptosis, and the second is reperfusion injury, which is also known as 'postresuscitation disease or syndrome' (Holzer 2010; Nolan 2008). Therapeutic hypothermia inhibits numerous pathways of these two mechanisms. In vivo ata comparing different starting points of therapeutic hypothermia are scarce, but there are a number of animal studies that have indicated that an earlier start of cooling, even in the resuscitation phase, might lead to an improved neurologic outcome (Abella 2008; Janata 2010; Kuboyama 1993; Sterz 1991).
Mild therapeutic hypothermia has been shown to be beneficial for neurologic outcome and survival for patients after resuscitation from sudden cardiac arrest (Arrich 2012). This evidence is mainly based on three randomized controlled trials. These randomized trials have initiated hypothermia as late as four hours after cardiac arrest. However, the question if an earlier cooling start may be more beneficial was raised in many experimental and human studies. It is important to note that 'earlier' refers to different time points of cooling, before resuscitation is started (Janata 2010; Zhao 2008), during resuscitation (Abella 2004; Yannopoulos 2009), immediately after resuscitation (Kuboyama 1993) (as opposed to a 15 minute delay), and after one hour (as opposed to a four hour delay) (Colbourne 1995). All these studies showed that an earlier initiation of cooling was more beneficial. However, a recent animal study demonstrated that there seemed to be no difference as long as the therapeutic hypothermia was established less than four hours after cardiac arrest (Che 2011). Also, retrospective human data show conflicting results. Some studies showed a better neurologic outcome with an earlier start of therapeutic hypothermia or achievement of therapeutic hypothermia (Sendelbach 2012; Wolff 2009) and others did not find that earlier hypothermia had a beneficial effect (Nielsen 2009). On the contrary, it has been shown in a cohort study that patients who reached the target temperature for mild therapeutic hypothermia early had poorer prognosis (Haugk 2011). This finding can probably be explained by the fact that patients with poor post-resuscitation conditions have compromised thermoregulation and less resistance against induction of hypothermia (Suffoletto 2009). This dilemma of differentiating a diagnostic sign (an early outcome) from an effective intervention can only be challenged by randomized trials. To date, a number of randomized controlled trials and pilot trials have been published, but to our knowledge there is no systematic review that has attempted to summarize their results.
Why it is important to do this review
Following the animal studies, a few case series have been published on the potential beneficial effect of an earlier versus a delayed start for therapeutic hypothermia with conflicting results (Nielsen 2009; Wolff 2009).
A few randomized controlled trials have been published that evaluate the feasibility and effectiveness of cooling during resuscitation and shortly after resuscitation versus in-hospital cooling, which is the standard care in most centres. The results were conflicting, partly due to the small study size of the pilot studies and maybe partly due to methodological shortcomings. Some showed a significant effect in the subgroups (Bernard 2003; Bernard 2010; Castren 2010; Kim 2007). To our knowledge there is no review or meta-analysis that has attempted to summarize the results of these studies.
Therapeutic hypothermia may improve neurologic outcome after resuscitation from sudden cardiac arrest even when its initiation is delayed. It is important to find out if timing of hypothermia initiation has an effect on clinical outcomes. If a prehospital start to cooling provides a further improvement for neurologic outcome, this effect should not be missed because of methodological shortcomings or small sample sizes of the preceding randomized controlled trials. If there is no further gain with a prehospital cooling start, the extra effort to start cooling in the field could be spared. This would help decision makers and guideline committees in advising healthcare professionals. For the scientific community, this will provide a basis on which to decide if further studies should be undertaken in this field and will serve as the basis for sample size calculations.
We aim to perform a systematic review and meta-analysis to assess the effectiveness of prehospital initiation of therapeutic hypothermia in comparison to later or in-hospital initiation of therapeutic hypothermia in patients with prehospital cardiac arrest. Our primary outcome measures, among other patient-relevant variables, are survival and neurologic outcome. We plan to undertake subgroup analyses according to the mode of therapeutic hypothermia and the cardiac arrest situation, and sensitivity analyses to assess the robustness of our estimates.
Criteria for considering studies for this review
Types of studies
We will include randomized and 'quasi-randomized' controlled trials. 'Quasi-randomized' refers to allocation procedures such as alternating days, odd and even days and the like, and these trials will be included because of the relative novelty of the intervention and the expected low number of trials available to date. We will exclude cross-over studies given the condition and the nature of the outcomes assessed.
Types of participants
We will include studies in adult patients with out-of-hospital cardiac arrest who received mild therapeutic hypothermia.
Types of interventions
We will compare prehospital induction of mild therapeutic hypothermia to in-hospital induction of mild therapeutic hypothermia.
Prehospital induction of mild therapeutic hypothermia (intervention): therapeutic hypothermia during resuscitation or shortly after resuscitation in the out-of-hospital setting.
Later induction of mild therapeutic hypothermia (standard therapy; control): in-hospital therapeutic hypothermia.
Types of outcome measures
- Survival. We will investigate short-term survival (closest to 30 days) and long-term survival (closest to six months).
- Neurologic outcome. Ideally we expect the outcome to be reported as best neurologic outcome during hospital stay and in cerebral performance categories (CPC) (Cummins 1991; Jennett 1975). Good neurological outcome is usually considered if the CPC is 1 or 2. If authors grouped this outcome along with other categories or cut-offs, or used other instruments for neurological outcome assessment like the modified Rankin Scale, we will accept this for our meta-analysis. In this case we will perform a sensitivity analysis based on the outcome definition.
- Adverse events, as reported by study authors
- Adverse events related to cooling methods, as reported by study authors
- Quality of life, as reported by study authors
- Length of stay in the intensive care unit (ICU) and in the hospital, as proxies for economic outcomes
Outcomes will not form part of the study eligibility assessment so that studies that meet the participant, intervention and comparison criteria will be included in the review even if they report no relevant outcomes.
Search methods for identification of studies
We aim to search the following databases (from inception to the current search date): the Cochrane Central Register of Controlled Trials (CENTRAL) in The Cochrane Library) (Appendix 1), MEDLINE (Appendix 2), EMBASE (Appendix 3), CINAHL (Appendix 4), PASCAL, and BIOSIS (Appendix 5). We will not apply any language restrictions.
We will use a search strategy for identifying randomized controlled trials (RCTs) in MEDLINE and EMBASE (Cochrane Handbook).
Searching other resources
In an attempt to identify further studies we will search the primary clinical trials registers accepted by the International Committee of Medical Journal Editors (see http://www.icmje.org/faq_clinical.html). We will also ask experts in the field whether they are aware of any ongoing, unpublished, or published trials on this subject. We will also search the reference lists of included studies and other reviews on the topic.
Data collection and analysis
Selection of studies
We will import all retrieved results into EndNote (version X5, Thomson Corporation) and eliminate any duplicates. Two authors (JA and CH) will independently scan titles and abstracts for relevance. For all references that are not excluded by both authors at this stage, we will independently retrieve the full texts and exam them for compliance with the inclusion criteria. We will link multiple reports of the same study. We will resolve discrepancies by discussion or by involving a third author as an arbiter.
Data extraction and management
Two authors will independently extract all relevant data into a predefined form (Appendix 6). We will resolve discrepancies by discussion or by involving a third author as arbiter. We will then enter data into the Cochrane Collaboration's software program Review Manager (RevMan 5.2).
Assessment of risk of bias in included studies
To assess the internal validity of the identified trials, we will use a domain-based evaluation, assessing random sequence generation, allocation concealment, blinding of outcome assessment (for neurological outcome and quality of life), incomplete outcome data (primary outcome), selective reporting (neurological outcome), exclusion of randomized patients from the analysis (primary outcome).
All relevant information will be extracted and tabulated independently by two review authors. We will resolve discrepancies by discussion or by involving a third author as arbiter.
We will assume that blinding of patients and personnel is not relevant, as patients are unconscious and cooling interventions are almost impossible to blind.
We will assess the domains as low risk of bias, unclear, or high risk of bias.
Measures of treatment effect
The primary measure of treatment effect for the primary outcomes will be the relative risk (risk ratio) for surviving and achieving good neurological recovery in patients allocated to prehospital initiation of hypothermia when compared to in-hospital initiation of hypothermia. The same applies for adverse events. For quality of life and length of stay data, differences will be used as measures of treatment effect. In the case that several instruments are used for quality of life assessment across studies, we will use standardized mean differences instead.
Unit of analysis issues
Cluster randomized trials
If cluster randomized trials should be included we will use estimates that allow for the clustered structure of data. In the absence of adequate estimates we will use appropriate approximations.
Studies with multiple treatment groups
In the case of multiple treatments (for example intra-arrest cooling versus prehospital arrest cooling versus in-hospital cooling) we will combine the groups to create a single pair-wise comparison, but avoid overall estimates.
Dealing with missing data
If a negligible amount of data are missing or convincingly missing at random, we will analyse only the available data. We will not employ data imputation or data replacement methods.
Assessment of heterogeneity
We will evaluate clinical and statistical heterogeneity. We will assess clinical heterogeneity by tabulating and informally inspecting relevant data. We will only perform quantitative synthesis if clinical heterogeneity is negligible. We will assess statistical heterogeneity using the I
Assessment of reporting biases
We will assess the presence of possible reporting bias using funnel plots (plotting the effect against precision) (Egger 1997; Cochrane Handbook) and inspecting them visually. If more than 10 studies are available we will use the Arcsine test (Rücker 2008) to formally test for funnel plot asymmetry.
We will calculate relative risks and their 95% confidence intervals for dichotomous variables. For continuous variables we will calculate mean differences or standardized mean differences depending on the comparability of measurement methods. Generally we will use random-effects models. We will use RevMan 5 for standard calculations and Stata 11 for meta-regression.
Subgroup analysis and investigation of heterogeneity
If possible, the following subgroups will be investigated.
- Efficacy of cooling methods in the intervention group and in the control group.
- Intra-arrest versus early post-arrest cooling in the intervention group.
- Duration of cardiac arrest.
- Primary cardiac rhythm.
- We will perform sensitivity analyses on the effect estimate by omitting studies with an overall 'high risk of bias' to assess the robustness of our estimates against within-study bias.
- If several methods of neurological outcome assessment are used we will investigate the robustness of the effect estimates.
Summary of findings
We will use the principles of the GRADE system (Guyatt 2008) to assess the quality of the body of evidence associated with the specific outcomes of survival and neurologic outcome in our review, and will construct a summary of findings (SOF) table using the GRADE software. The GRADE approach appraises the quality of a body of evidence based on the extent to which one can be confident that an estimate of effect or association reflects the item being assessed. The quality of a body of evidence considers within study risk of bias (methodologic quality), the directness of the evidence, heterogeneity of the data, precision of effect estimates, and risk of publication bias.
We would like to thank Dr Nicola Petrucci (content editor), Dr Cathal Walsh (statistical editor), Dr Kjetil Sunde, Dr Clifton Callaway, Dr Jasmeet Soar (peer reviewers), and Jane Cracknell (Cochrane Anaesthesia Review Group Managing Editor) for their help and editorial advice during the preparation of the protocol for the systematic review.
Appendix 1. Search strategy: CENTRAL, The Cochrane Library
#1 MeSH descriptor Resuscitation explode all trees
#2 MeSH descriptor Cardiopulmonary Resuscitation explode all trees
#3 MeSH descriptor Resuscitation Orders explode all trees
#4 MeSH descriptor Heart Arrest explode all trees
#5 MeSH descriptor Heart Massage explode all trees
#6 ((cardio?pulmonary or order*) near2 resuscitation):ti,ab
#8 ((circulatory or circulation or cardiac) near arrest):ti,ab or heart standstill:ti,ab
#9 (#1 OR #2 OR #3 OR #4 OR #5 OR #6 OR #7 OR #8)
#10 MeSH descriptor Cryotherapy explode all trees
#11 MeSH descriptor Hypothermia explode all trees
#12 MeSH descriptor Hypothermia, Induced explode all trees
#13 ((resuscitative or therapeutic or artificial or induced or extracorporeal) near hypothermia)
#14 artificial hibernation or body cooling or refrigeration anesthesia or body temperature:ti,ab or refrigeration:ti,ab
#15 (#10 OR #11 OR #12 OR #13 OR #14)
#16 (#9 AND #15)
Appendix 2. Search strategy: MEDLINE (OvidSP)
1. Resuscitation/ or Cardiopulmonary Resuscitation/ or Resuscitation Orders/ or Heart Arrest/ or Heart Massage/ or advanced cardiac life support.mp. or ((cardio?pulmonary or order*) adj2 resuscitation).ti,ab. or reanimation.ti,ab. or ((circulatory or circulation or cardiac) adj3 arrest).ti,ab. or heart standstill.ti,ab.
2. Cryotherapy/ or Hypothermia/ or Circulatory Arrest, Deep Hypothermia Induced/ or Hypothermia, Induced/ or ((resuscitative or therapeutic or artificial or induced or extracorporeal) adj3 hypothermia).mp. or artificial hibernation.mp. or body cooling.mp. or chilling.mp. or refrigeration anesthesia.mp. or body temperature.ti,ab. or refrigeration.ti,ab.
3. 1 and 2
4. ((randomised controlled trial or controlled clinical trial).pt. or randomised.ab. or placebo.ab. or clinical trials as topic.sh. or randomly.ab. or trial.ti.) not (animals not (humans and animals)).sh.
5. 3 and 4
Appendix 3. Search strategy: EMBASE (OvidSP)
1. resuscitation/ or heart arrest/ or heart massage/ or advanced cardiac life support.mp. or ((cardio?pulmonary or order*) adj2 resuscitation).ti,ab. or reanimation.ti,ab. or ((circulatory or circulation or cardiac) adj3 arrest).ti,ab. or heart standstill.ti,ab.
2. cryotherapy/ or hypothermia/ or ((resuscitative or therapeutic or artificial or induced or extracorporeal) adj3 hypothermia).mp. or artificial hibernation.mp. or body cooling.mp. or chilling.mp. or refrigeration anesthesia.mp. or body temperature.ti,ab. or refrigeration.ti,ab.
3. 1 and 2
4. (placebo.sh. or controlled study.ab. or random*.ti,ab. or trial*.ti,ab. or ((singl* or doub* or trebl* or tripl*) adj3 (blind* or mask*)).ti,ab.) not (animals not (humans and animals)).sh.
5. 3 and 4
Appendix 4. Search strategy: CINAHL (EBSCOhost)
S1 ( (MH "Resuscitation") OR (MH "Resuscitation Orders") OR (MH "Resuscitation, Cardiopulmonary") OR (MH "Heart Arrest") OR (MH "Heart Massage") ) OR AB ( ((cardio?pulmonary or order*) and resuscitation) ) OR AB reanimation OR ( (circulatory or circulation or cardiac) and arrest ) OR heart standstill
S2 ( (MH "Cryotherapy") OR (MH "Hypothermia") OR (MH "Hypothermia, Induced") ) OR ( ((resuscitative or therapeutic or artificial or induced or extracorporeal) and hypothermia) ) OR artificial hibernation OR body cooling OR refrigeration anesthesia
S3 ( (MH "randomised Controlled Trials") OR (MH "Random Assignment") OR (MH "Prospective Studies") OR (MH "Multicenter Studies") OR (MH "Clinical Trials") OR (MH "Clinical Trial Registry") OR (MH "Double-Blind Studies") OR (MH "Single-Blind Studies") OR (MH "Triple-Blind Studies") OR (MH "Placebos") ) OR ( random* or controlled clinical trial or placebo )
S4 S1 and S2 and S3
Appendix 5. Search Strategy: BIOSIS (OvidSP) and PASCAL
1. advanced cardiac life support.mp. or ((cardio?pulmonary or order*) adj2 resuscitation).ti,ab. or reanimation.ti,ab. or ((circulatory or circulation or cardiac) adj3 arrest).ti,ab. or heart standstill.ti,ab.
2. (((resuscitative or therapeutic or artificial or induced or extracorporeal) adj3 hypothermia) or artificial hibernation or body cooling or chilling or refrigeration anesthesia).mp. or body temperature.ti,ab. or refrigeration.ti,ab.
3. 1 and 2
Appendix 6. Data extraction form
Secondary outcomes (dichotomous):
Secondary outcomes (continuous):
Contributions of authors
Jasmin Arrich: (JA), Christof Havel (CH), Michael Holzer (MH), Harald Herkner (HH)
Conceiving the review: JA, CH, MH, and HH
Co-ordinating the review: HH and JA
Undertaking manual searches: JA, HH, and CH
Screening search results: CH and JA
Organizing retrieval of papers: JA
Screening retrieved papers against inclusion criteria: CH, HH and JA
Appraising quality of papers: JA, CH, MH, and HH
Abstracting data from papers: JA and CH
Writing to authors of papers for additional information: JA
Providing additional data about papers: JA
Obtaining and screening data on unpublished studies: JA, CH, MH and HH
Data management for the review: HH and JA
Entering data into Review Manager (RevMan 5.2): JA and CH
RevMan statistical data: HH and JA
Other statistical analysis not using RevMan: HH
Interpretation of data: JA, CH, MH and HH
Statistical inferences: HH
Writing the review: JA, CH, MH and HH
Securing funding for the review: not applicable
Performing previous work that was the foundation of the present study: JA, CH, MH and HH
Guarantor for the review (one author): JA
Person responsible for reading and checking review before submission: JA
Declarations of interest
Jasmin Arrich received travel grants for scientific conferences from EMCOOLS and Euromed.
Michael Holzer received travel grants for scientific conferences and honoraria for lectures from EMCOOLS and provided consultancy for Leerink Swann LLC. He received grants for scientific studies from Velomedix Inc. The Department of Emergency Medicine of the Medical University of Vienna received a research grant from Velomedix Inc. and will receive a research grant form Philips in the near future.
Michael Holzer and Christof Havel were involved in the design, conduct and publication of the HACA 2002 (Hypothermia 2002) trial.
Harald Herkner has no conflict of interest.
Sources of support
- None, Not specified.
- No sources of support supplied