Effect of Lower vs Higher Oxygen Saturation Targets on Survival to Hospital Discharge Among Patients Resuscitated After Out-of-Hospital Cardiac Arrest: The EXACT Randomized Clinical Trial

Stephen A Bernard; Janet E Bray; Karen Smith; Michael Stephenson; Judith Finn; Hugh Grantham; Cindy Hein; Stacey Masters; Dion Stub; Gavin D Perkins; Natasha Dodge; Catherine Martin; Sarah Hopkins; Peter Cameron

doi:10.1001/jama.2022.17701

. 2022 Oct 26;328(18):1818–1826. doi: 10.1001/jama.2022.17701

Effect of Lower vs Higher Oxygen Saturation Targets on Survival to Hospital Discharge Among Patients Resuscitated After Out-of-Hospital Cardiac Arrest

The EXACT Randomized Clinical Trial

Stephen A Bernard ^1,^2,^3,^✉, Janet E Bray ^1,^3,⁴, Karen Smith ^1,^2,⁵, Michael Stephenson ^1,^2,⁵, Judith Finn ^1,⁴, Hugh Grantham ^4,^6,⁷, Cindy Hein ⁸, Stacey Masters ⁴, Dion Stub ^1,^2,³, Gavin D Perkins ⁹, Natasha Dodge ¹, Catherine Martin ¹⁰, Sarah Hopkins ², Peter Cameron ^1,³, for the EXACT Investigators

¹Department of Epidemiology and Preventive Medicine, Monash University, Melbourne, Victoria, Australia

²Ambulance Victoria, Melbourne, Victoria, Australia

³Alfred Hospital, Melbourne, Victoria, Australia

⁴Prehospital, Resuscitation and Emergency Care Research Unit, Curtin University, Perth, Western Australia, Australia

⁵Department of Paramedicine, Monash University, Melbourne, Victoria, Australia

⁶SA Ambulance Service, Adelaide, South Australia, Australia

⁷Flinders Medical Centre, Adelaide, South Australia, Australia

⁸Flinders University, Adelaide, South Australia, Australia

⁹Warwick University, Warwick, United Kingdom

¹⁰Monash University, Data Science and AI Platform, Melbourne, Victoria, Australia

Group Information: The EXACT Investigators appear in Supplement 4.

^✉

Corresponding Author: Stephen Bernard, MD, Ambulance Victoria, 375 Manningham Rd, Doncaster, Victoria 3108, Australia (stephen.bernard@ambulance.vic.gov.au).

Accepted for Publication: September 9, 2022.

Published Online: October 26, 2022. doi:10.1001/jama.2022.17701

Author Contributions: Drs Bernard and Bray had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Drs Bernard and Bray contributed equally to this article as co–first authors.

Concept and design: Bernard, Bray, Smith, Stephenson, Finn, Grantham, Hein, Stub, Perkins, Cameron.

Acquisition, analysis, or interpretation of data: Bray, Smith, Stephenson, Grantham, Hein, Masters, Stub, Perkins, Dodge, Martin, Hopkins, Cameron.

Drafting of the manuscript: Bernard, Bray.

Critical revision of the manuscript for important intellectual content: Smith, Stephenson, Finn, Grantham, Hein, Masters, Stub, Perkins, Dodge, Martin, Hopkins, Cameron.

Statistical analysis: Martin.

Obtained funding: Bernard, Bray, Smith, Stephenson, Finn, Hein, Stub, Perkins, Cameron.

Administrative, technical, or material support: Bray, Smith, Stephenson, Grantham, Hein, Masters, Stub, Dodge, Hopkins, Cameron.

Supervision: Bernard, Bray, Smith, Stephenson, Perkins, Cameron.

Conflict of Interest Disclosures: None reported.

Funding/Support: The trial was funded by a project grant from the National Health and Medical Research Council (NHMRC; project grant number APP1107509). The Australasian Resuscitation Outcomes Consortium and the NHMRC Prehospital Emergency Care Centre of Research Excellence (number 116453) provided administrative support. Dr Finn receives an NHMRC investigator grant (number 1174838), Dr Cameron receives a Medical Research Future Fund fellowship (number 1139686), and Drs Bray and Stub receive Heart Foundation fellowships (numbers 104751/105793). Dr Perkins is supported by the National Institute for Health Research Applied Research Collaboration West Midlands. Drs Bernard, Bray, Smith, Stephenson, Finn, Grantham, Hein, Stub, Perkins, and Cameron received funding from the NHMRC.

Role of the Funder/Sponsor: The NHMRC 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.

Group Information: The EXACT Investigators appear in Supplement 4.

Disclaimer: The views expressed are those of the author(s) and not necessarily those of the National Institute for Health and Care Research or the Department of Health and Social Care.

Data Sharing Statement: See Supplement 5.

Meeting Presentation: This paper was presented at the European Society of Intensive Care Medicine meeting; October 26, 2022; Paris, France.

Additional Contributions: We thank the paramedics at Ambulance Victoria and SA Ambulance Service for their contribution in screening, enrolling, and treating study patients. We also thank members of the data and safety monitoring committee (Christopher Reid, PhD [Curtin University]; Laurent Billot, MSc [University of New South Wales]; and Ian Patrick, ASM [Monash University]); and Stuart Howell, PhD (Monash University), for providing data to this committee with compensation for this role.

^✉

Corresponding author.

PMCID: PMC9608019 PMID: 36286192

Key Points

Question

Among patients with return of spontaneous circulation after out-of-hospital cardiac arrest, does targeting an oxygen saturation of 90% to 94%, compared with 98% to 100%, until admission to the intensive care unit improve survival?

Findings

In this randomized clinical trial that included 425 patients and was stopped early due to the COVID-19 pandemic, targeting an oxygen saturation of 90% to 94%, compared with an oxygen saturation of 98% to 100%, did not significantly improve survival to hospital discharge (rates of survival to hospital discharge, 38.3% vs 47.9%).

Meaning

The findings do not support use of an oxygen saturation target of 90% to 94% in the out-of-hospital setting after resuscitation from cardiac arrest.

Abstract

Importance

The administration of a high fraction of oxygen following return of spontaneous circulation in out-of-hospital cardiac arrest may increase reperfusion brain injury.

Objective

To determine whether targeting a lower oxygen saturation in the early phase of postresuscitation care for out-of-hospital cardiac arrest improves survival at hospital discharge.

Design, Setting, and Participants

This multicenter, parallel-group, randomized clinical trial included unconscious adults with return of spontaneous circulation and a peripheral oxygen saturation (Spo₂) of at least 95% while receiving 100% oxygen. The trial was conducted in 2 emergency medical services and 15 hospitals in Victoria and South Australia, Australia, between December 11, 2017, and August 11, 2020, with data collection from ambulance and hospital medical records (final follow-up date, August 25, 2021). The trial enrolled 428 of a planned 1416 patients.

Interventions

Patients were randomized by paramedics to receive oxygen titration to achieve an oxygen saturation of either 90% to 94% (intervention; n = 216) or 98% to 100% (standard care; n = 212) until arrival in the intensive care unit.

Main Outcomes and Measures

The primary outcome was survival to hospital discharge. There were 9 secondary outcomes collected, including hypoxic episodes (Spo₂ <90%) and prespecified serious adverse events, which included hypoxia with rearrest.

Results

The trial was stopped early due to the COVID-19 pandemic. Of the 428 patients who were randomized, 425 were included in the primary analysis (median age, 65.5 years; 100 [23.5%] women) and all completed the trial. Overall, 82 of 214 patients (38.3%) in the intervention group survived to hospital discharge compared with 101 of 211 (47.9%) in the standard care group (difference, −9.6% [95% CI, −18.9% to −0.2%]; unadjusted odds ratio, 0.68 [95% CI, 0.46-1.00]; P = .05). Of the 9 prespecified secondary outcomes collected during hospital stay, 8 showed no significant difference. A hypoxic episode prior to intensive care was observed in 31.3% (n = 67) of participants in the intervention group and 16.1% (n = 34) in the standard care group (difference, 15.2% [95% CI, 7.2%-23.1%]; OR, 2.37 [95% CI, 1.49-3.79]; P < .001).

Conclusions and Relevance

Among patients achieving return of spontaneous circulation after out-of-hospital cardiac arrest, targeting an oxygen saturation of 90% to 94%, compared with 98% to 100%, until admission to the intensive care unit did not significantly improve survival to hospital discharge. Although the trial is limited by early termination due to the COVID-19 pandemic, the findings do not support use of an oxygen saturation target of 90% to 94% in the out-of-hospital setting after resuscitation from cardiac arrest.

Trial Registration

ClinicalTrials.gov Identifier: NCT03138005

This study examines whether reducing oxygen fraction following resuscitation from out-of-hospital cardiac arrest improves survival at hospital discharge.

Introduction

Determination of an optimal postarrest oxygen target is a recognized gap in existing knowledge.¹ Based on data from Australia from 2014 to 2016, the majority of patients with out-of-hospital cardiac arrest (OHCA) who achieve return of spontaneous circulation (ROSC) remain in a comatose state and require assisted ventilation during transport to the hospital (approximately 80%).² Standard out-of-hospital practice is to provide a fraction of inspired oxygen (Fio₂) of 100% until arrival at the emergency department, where the Fio₂ is adjusted according to local mechanical ventilation and postarrest care protocols.

However, animal data³ and clinical studies^4,5 have indicated that the administration of 100% oxygen during the early postarrest period may cause hyperoxia and associated neurological injury and less favorable clinical outcomes. The mechanism of harm of hyperoxia is thought to be related to an increase in the production of oxygen free radical molecules that are known to injure neurons (reperfusion injury).⁶ Supplemental oxygen may also cause additional injury to the myocardium in patients with coronary artery occlusion.⁷ Titration of oxygen in the out-of-hospital setting is feasible⁸; however, the efficacy and safety of this approach are uncertain.⁹

The Reduction of Oxygen After Cardiac Arrest (EXACT) trial was conducted to determine whether reducing oxygen fraction following resuscitation from OHCA in the out-of-hospital setting to target a peripheral oxygen saturation (Spo₂) of 90% to 94%, compared with a target of 98% to 100%, improves survival at hospital discharge.

Methods

Trial Design

The EXACT trial was an investigator-initiated, multicenter, parallel-group, randomized clinical trial with patients, statistician, and 12-month outcome assessors blinded to treatment randomization.¹⁰ The study was conducted in 2 emergency medical services (EMS) and 15 hospitals in 2 Australian states. The protocol (Supplement 1), developed by the study steering committee, was approved by human research ethics committees at each hospital. The statistical analysis plan was developed by a statistician and the study steering committee (Supplement 2). Differences between the trial protocol, statistical analysis plan, and this report are described in eAppendix 1 in Supplement 3. An independent data and safety monitoring committee periodically reviewed blinded efficacy and safety data, with the option to request unblinded data if required. In Victoria, patients were enrolled under a waiver of consent, with survivors or surrogates notified of enrollment with the option to opt-out of data collection. In South Australia, patients were enrolled under a waiver of consent, with consent received for all patients who survived. The EMS and relevant practices are described in eAppendix 2 in Supplement 3 and resuscitation treatment followed the Australian Resuscitation Guidelines.¹¹

Patients

Patients with ROSC following OHCA of presumed cardiac cause were eligible. Inclusion criteria were age 18 years or older, unconscious after ROSC, advanced airway (endotracheal tube or supraglottic airway), Spo₂ of at least 95% while receiving more than 10 L/min of oxygen or Fio₂ of 100%, and transport planned to a participating hospital. Paramedics determined the presumed cause at the time of enrollment, and patients with no other obvious cause of arrest were presumed to be of cardiac etiology.^12,13

Patients were excluded if, at the time of enrollment, they had an obvious noncardiac cause of arrest (ie, respiratory, trauma, hanging, drowning), were known to or suspected of being pregnant, were dependent on others for activities of daily living or had a do-not-resuscitate order, or were receiving home oxygen therapy.

Randomization

Patients were screened and randomized by paramedics trained in the study protocol. Patients were randomized in a 1:1 ratio to have a targeted Spo₂ of 90% to 94% (intervention) or 98% to 100% (standard care). Randomization was generated in blocks of 10 by a computer-generated code and stratified by EMS. Paramedic teams were provided with sequentially numbered blocks of 10 sealed opaque envelopes containing a card and an airway tag indicating treatment randomization.

Intervention

In the intervention group, oxygen was initially reduced to 4 L/min via an oxygen reservoir bag (Fio₂ of approximately 0.7), or an oxygen/air mix setting (Fio₂ of approximately 0.6) if the patient was receiving mechanical ventilation, and then titrated to maintain an Spo₂ of 90% to 94%. Patients randomized to receive standard care were administered high-flow oxygen (Fio₂ of 100% if receiving mechanical ventilation or >10 L/min of oxygen via an oxygen reservoir bag) in the out-of-hospital setting and then underwent oxygen titration via a ventilator to maintain an Spo₂ of 98% to 100% in the hospital. The intervention continued until the first arterial blood gas measurement in the intensive care unit (ICU). Titration to 100% oxygen was allowed for subsequent intubation or hypoxic events (Supplement 1). Treatment of patients after hospital arrival followed the hospital’s standard practices for OHCA management, and hospital clinicians were not blinded to treatment randomization.

Data Collection

Unblinded research staff collected data from EMS and hospital medical records. The 12-month outcomes were obtained using telephone interview of patients or their proxies by trained staff who were blinded to treatment randomization.

Outcomes

The primary outcome was survival to hospital discharge. Prespecified secondary outcomes collected prior to hospital discharge (Supplement 2 and eAppendix 1 in Supplement 3) were rates of rearrest and hypoxia (Spo₂ <90%) before ICU admission, myocardial injury (peak troponin level), survival to ICU discharge, ICU and hospital length of stay, cause of in-hospital mortality, favorable neurological outcome (Cerebral Performance Category score of 1-2) at hospital discharge, and discharge destination (in survivors). The Cerebral Performance Category score is a 5-point scale measuring neurological outcomes following brain injury, ranging from good functioning (score of 1 to 2) to brain death (score of 5).¹⁴

Prespecified secondary outcomes measured at 12 months included survival and health-related quality of life. Measures used were the mental and physical components of the generic 12-Item Short Form Health Survey, with higher scores representing greater mental and physical health and scores greater than 50 representing no disability¹⁵; the EuroQol 5 Dimension 5 index, which indexes a person’s self-rated health, ranging from −0.59 to 1, with the lowest score indicating the worst imaginable health state and the highest indicating the best imaginable health state¹⁶; the Glasgow Outcome Scale-Extended, with scores categorized into good recovery (score of 1-2), moderate disability (score of 3-4), or severe disability to death (score of 5-8)¹⁷; and the modified Rankin Scale score, a 7-point scale measuring the degree of disability and dependence in daily living, with favorable outcomes considered to be scores of 0 to 2 (no or minimal disability) and unfavorable outcomes defined as scores of 3 to 5 (significant disability) or 6 (death).¹⁸

Adverse Events

The predefined serious adverse events were sustained hypoxia (Spo₂ <90%), unresponsive to 100% oxygen, and rearrest in the setting of hypoxia (Spo₂ <90%).

Sample Size

The study was planned to enroll 1416 patients, which would have allowed 90% power to determine a relative difference of 25% between the groups for the primary outcome. The standard care group was expected to have a 35% survival rate at hospital discharge¹⁹ and the intervention group was anticipated to have a 44% survival rate. The effect size was based on a large observational clinical study²⁰ and 2 meta-analyses of observational studies.^21,22 After adjusting for the interim analysis, the study would have required 643 patients per group with 90% power and restricted α = .049, with 10% added to this sample size to account for loss to follow-up.

Statistical Analysis

Groups were analyzed according to randomization group, excluding patients who did not consent (South Australia only) or requested data be withdrawn. A secondary prespecified analysis of all outcomes was performed excluding patients enrolled with known exclusion criteria.

As a result of the early cessation of the study, the statistical plan was adjusted prior to database lock to remove the statistical comparison for 12-month outcomes and to include a multivariable analysis of the primary outcome as a supplementary analysis (Supplement 2 and eAppendix 1 in Supplement 3). Primary and secondary outcomes were analyzed by a statistician blinded to treatment randomization, and all data related to oxygenation were withheld until that analysis was complete.

Binary and categorical variables are expressed as proportions and percentages and continuous data are summarized as mean and SD or as median and IQR as appropriate. Per the published protocol,¹⁰ the analysis of the primary outcome and other binary or categorical outcomes was tested using χ² tests. Estimates and their 95% CIs were determined using logistic regression. Normally distributed data were tested using t tests and nonnormal continuous data were analyzed using Wilcoxon rank sum tests, except for length of stay outcomes, for which median regression was performed. There was no imputation performed for other missing data.

Planned subgroup analyses for the primary outcome were assessed with the use of regression models with tests for the interactions between the subgroup and the randomized group. Per the statistical analysis plan, subgroups included were age (<65 y), sex, witnessed arrest, bystander cardiopulmonary resuscitation (CPR), shockable rhythm, time from collapse to ROSC, and ST-segment elevation myocardial infarction. Due to small numbers and data collection, 2 additional prespecified subgroups were not analyzed (eAppendix 1 in Supplement 3).

Exploratory post hoc analyses are detailed in eAppendix 1 in Supplement 3. This analysis included a multivariable logistic regression for survival to hospital discharge by treatment group, adjusting for known predictors of survival¹³ and hospital site using mixed-effects logistic regression. Additional post hoc analyses were performed to examine differences between treatment groups for survival to discharge stratified by the time from emergency call to randomization, the timing of withdrawal of life-sustaining treatment, and rates of hypoxia and rearrest without ROSC in the subgroup of patients who received bystander CPR.

Subgroup analyses and secondary end points were not adjusted for multiplicity. Because of the potential for type I error due to multiple comparisons, findings for analyses of secondary outcomes and subgroups should be interpreted as exploratory. Reported P values are 2-sided, with a P value of <.05 considered significant. The statistical software used was STATA, version 16.0 (Stata Corporation).

Results

On August 7, 2020, the study steering committee discussed and approved ceasing the trial based on decreased trial enrollment and COVID-19 pandemic–related changes in OHCA protocols and survival rates.²³ The ethics committees, paramedics, and site investigators were notified and the trial ceased on August 11, 2020, at which time 428 patients had been enrolled. This decision was made without knowledge of the outcomes of the study, per the CONSERVE (CONSORT and SPIRIT Extension for RCTs Revised in Extenuating Circumstances) 2021 statement.²⁴ Similar numbers of unused randomization envelopes were accounted for in each study group (47% in the intervention group and 49% in standard care group).

Patient Characteristics

Between December 11, 2017, and August 11, 2020, a total of 425 of 428 patients were enrolled and included in the main analysis, with 214 randomized to the intervention group (titrated to maintain Spo₂ of 90%-94%) and 211 randomized to the standard care group (titrated to maintain Spo₂ of 98%-100%) (Figure 1). Three patients who were randomized were not included in the main analyses (consent was not received for 2 and data withdrawal requested for 1). There were 32 ineligible patients randomized (18 in the intervention group and 14 in the standard care group; eTable 1 in Supplement 3). Follow-up at hospital discharge was complete for all patients.

The baseline data were similar between the 2 groups (Table 1). The median (IQR) age was 65.5 (53.1-76.4) years and 100 patients (23.5%) were women. Most arrests were bystander-witnessed (74.5%), most individuals received bystander CPR (81.0% of bystander-witnessed arrests), and most were initially in a shockable cardiac rhythm (62.0%).

Table 1. Baseline Characteristics in a Study of the Effect of Lower vs Higher Oxygen Saturation (Spo₂) Targets on Survival to Hospital Discharge After Cardiac Arrest.

Characteristics	No. (%)
Characteristics	Target Spo₂ 90%-94% (n = 214)	Target Spo₂ 98%-100% (n = 211)
Age, median (IQR), y	66.4 (56.8-77.5)	64.2 (50.1-74.9)
Sex
Men	163 (76.2)	162 (76.8)
Women	51 (23.8)	49 (23.2)
State
Victoria	207 (96.7)	195 (92.3)
South Australia	7 (3.3)	16 (7.7)
Arrest witnessed^a
No	48 (22.5)	42 (20.0)
Yes
Bystander	155 (72.8)	160 (76.2)
EMS	10 (4.7)	8 (3.8)
Bystander CPR^b^,^c	163 (81.5)	157 (80.1)
Location of cardiac arrest
Private residence	147 (68.7)	138 (65.4)
Public place	66 (30.8)	69 (32.7)
Other	1 (0.5)	4 (1.9)
Presumed cause
Cardiac	199 (93.0)	206 (97.6)
Other	15 (7.0)	5 (2.4)
EMS response time, median (IQR), min^c	8.0 (6.0-11.0)	7.0 (6.0-10.0)
First arrest rhythm^d
Shockable	128 (60.2)	135 (64.0)
Asystole	29 (13.6)	28 (13.3)
Pulseless electrical activity	50 (23.4)	44 (20.9)
Unknown unshockable	6 (2.8)	4 (1.9)
No. of defibrillations, median (IQR)	1 (0.0-4.0)	2 (0.0-4.0)
Epinephrine IV administration (intra-arrest)	167 (78.0)	163 (77.3)
Epinephrine IV (intra-arrest) dose, mean (SD), mg^e	3.1 (2.2)	2.9 (2.0)
Time, median (IQR), min
Total downtime (call time to ROSC)	27.0 (20.0-37.0)	25.0 (18.0-34.0)
EMS arrival to ROSC^c	18.0 (11.0-27.5)	16.0 (11.0-25.0)
ROSC to randomization	36.0 (25.0-48.0)	36.0 (23.0-46.0)
ROSC to ED^e	57.0 (45.0-74.0)	57.0 (47.0-70.0)
Post-ROSC drugs
Midazolam	199 (93.0)	195 (92.4)
Morphine	183 (85.5)	171 (81.0)
Fentanyl	81 (37.9)	91 (43.1)
Ketamine^e	11 (5.1)	9 (4.3)
Supraglottic airway	170 (79.4)	169 (80.1)
endotracheal tube inserted	201 (93.9)	192 (91.0)
Depolarizing muscle relaxant	63 (29.4)	63 (29.9)
Nondepolarizing muscle relaxant	162 (75.7)	165 (78.2)
Prehospital mechanical ventilation	157 (73.7)	159 (75.7)
STEMI on 12-lead ECG before the hospital	68 (31.8)	64 (30.3)

Open in a new tab

Abbreviations: CPR, cardiopulmonary resuscitation; ECG, electrocardiogram; ED, emergency department; EMS, emergency medical services; ETT, endotracheal tube; ROSC, return of spontaneous circulation; STEMI, ST-segment elevation myocardial infarction.

^{^a}

Data were missing for 1 participant in the 98%-100% group and 1 participant in the 90%-94% group.

^{^b}

Data were missing for 7 participants in the 98%-100% group and 3 participants in the 90%-94% group.

^{^c}

Data exclude EMS-witnessed cases.

^{^d}

Data were missing for 1 participant in the 90%-94% group.

^{^e}

Data were missing for 1 participant in the 98%-100% group.

The median (IQR) time from ROSC to randomization was 36 (23-46) minutes. Group comparisons for the pulse oximeter data for the 6 hours following randomization are shown in eFigure 1 in Supplement 3, time to first arterial blood gas measurement is shown in eTable 2 in Supplement 3, and oxygenation and oxygen delivery during the intervention period are shown in Table 2. These data show differences between groups for oxygenation and oxygen delivery in line with the trial protocol. For example, the median (IQR) last recorded Fio₂ in the emergency department was 0.5 (0.4-0.8) in the intervention group and 0.8 (0.5-1.0) in the standard care group.

Table 2. Clinical Measures in a Study of the Effect of Lower vs Higher Oxygen Saturation Targets (Spo₂) on Survival to Hospital Discharge After Cardiac Arrest.

Clinical parameter	Median (IQR)
Clinical parameter	Target Spo₂ 90%-94% (n = 214)	Target Spo₂ 98%-100% (n = 211)
Prehospital
Spo₂
At randomization	99 (97-100)	99 (97-100) [n = 209]
At ED arrival^a	97 (94-99)	99 (98-100) [n = 209]
ED^a
Oxygen flow rate on ED arrival, L/min	2 (2-4)	12 (10-15)
Fio₂ on ED arrival, No. (%)
<0.8	134 (84.8)	4 (2.2)
≥0.8	24 (15.2)	175 (97.8)
Last recorded Fio₂ in ED	0.5 (0.4-0.8)	0.8 (0.5-1.0)
First pH in ED	7.17 (7.07-7.24)	7.17 (7.07-7.25)
First Paco₂ in ED, mm Hg	60 (51-70)	60 (51-71)
First PaO₂ in ED (arterial blood gas), mm Hg	95 (74-138)	130 (89-237)
Spo₂ in ED
First	96 (93-99)	99 (97-100)
Last	97 (94-100)	99 (97-100)
ICU and hospital ^a ^, ^b
First pH in ICU	7.26 (7.18-7.32)	7.26 (7.20-7.34)
First Paco₂ in ICU, mm Hg	45 (38-52)	46 (38-52)
First Pao₂ in ICU, mm Hg	99 (73-146)	114 (83-177)
First Spo₂ in ICU, %	98 (95-100)	99 (97-100)
Target temperature control, No. (%)	n = 190	n = 186
No	47 (24.5)	35 (17.8)
33-36 °C	121 (63.0)	138 (70.5)
>36 °C	24 (12.5)	24 (12.1)
Coronary angiogram, No. (%)	128 (60.4) [n = 212]	127 (60.2) [n = 211]
PCI	72 (56.2) [n = 128]	63 (50.0) [n = 127]
Coronary artery bypass graft surgery, No. (%)	3 (1.4)	5 (2.4) [n = 210]

Open in a new tab

Abbreviations: ED, emergency department; ICU, intensive care unit; PCI, percutaneous coronary intervention.

^{^a}

Data exclude 1 participant who died before hospital admission.

^{^b}

Data for patients admitted to the ICU only: 192 in the intervention group and 197 in the standard care group.

Other clinical measures and postarrest treatments are shown in Table 2. Similar rates for key elements of postresuscitation care were seen between the 2 groups, with 60.3% undergoing coronary angiography and 78.9% of those admitted to the ICU receiving targeted temperature control.

Outcomes

Primary Outcome

The number of patients who survived to hospital discharge was 82 of 214 (38.3%) in the intervention group compared with 101 of 211 (47.9%) in the standard care group (difference, −9.6% [95% CI, −18.9% to −0.2%]; odds ratio [OR], 0.68 [95% CI, 0.46-1.00]; P = .05) (Table 3). Survival to hospital discharge, in a sensitivity analysis restricted to eligible patients (n = 393), was not statistically different between groups (OR, 0.71 [95% CI, 0.48-1.06]; P = .09) (eTable 3 in Supplement 3).

Table 3. Outcomes Collected During Intensive Care Unit and Hospital Stay.

Outcome	No. (%)		Difference (95% CI)^a	Odds ratio (95% CI)	P value^b
Outcome	Target Spo₂ 90%-94% (n = 214)	Target Spo₂ 98%-100% (n = 211)	Difference (95% CI)^a	Odds ratio (95% CI)	P value^b
Primary
Survival to hospital discharge	82 (38.3)	101 (47.9)	−9.6 (−18.9 to −0.2)	0.68 (0.46 to 1.00)	.05
Secondary
Rearrest
Pre-ICU^c	27 (12.7) [n = 213]	21 (10.0) [n = 209]	2.6 (−3.4 to 8.7)	1.30 (0.71 to 2.38)	.40
Prehospital	7 (3.3)	3 (1.4)	1.8 (−1.0 to 4.7)
ED	26 (12.2) [n = 213]	20 (9.5) [n = 210]	2.7 (−3.2 to 8.6)
Hypoxia (any Spo₂ <90%) prior to ICU	67 (31.3)	34 (16.1)	15.2 (7.2 to 23.1)	2.37 (1.49 to 3.79)	<.001
Peak troponin, median (IQR)	n = 193	n = 198
Troponin T	581 (134 to 2363)	557 (179 to 2234)	24 (−405 to 453)		.91
Troponin I	1838 (316 to 8578)	1550 (270 to 6710)	288 (−756 to 1332)		.59
Survival to ICU discharge	96/192 (50.0)	106/197 (53.8)	−3.8 (−13.7 to 6.1)	0.86 (0.58 to 1.28)	.45
ICU length of stay, median (IQR), d
Survivors	4.0 (2.0 to 6.0) [n = 96]	4.0 (2.0 to 6.0) [n = 105]	0.0 (−1.2 to 1.2)		>.99
Deaths	3.0 (1.0 to 5.5) [n = 96]	4.0 (1.0 to 7.0) [n = 91]	−1.0 (−2.8 to 0.8)		.27
Hospital length of stay, median (IQR), d
Survivors	11.0 (7.0 to 17.0) [n = 82]	11.0 (7.0 to 16.0) [n = 101]	0.0 (−2.8 to 2.8)		>.99
Nonsurvivors	3.0 (1.0 to 6.0) [n = 132]	4.0 (1.0 to 7.0) [n = 109]	−1.0 (−2.4 to 0.4)		.16
Cause of in-hospital mortality, No. (%)
Rearrest with no ROSC	14 (10.8)	7 (6.4)	4.4 (−2.7 to 11.4)	1.77 (0.67 to 4.63)	.25
Treatment withdrawn
Hypoxic brain injury	77 (59.2)	68 (62.4)	−3.2 (−15.6 to 9.3)	1 [Reference]
Cardiogenic shock	14 (10.8)	11 (10.1)	0.7 (−7.1 to 8.5)	1.12 (0.48 to 2.64)	.79
Palliation	2 (1.5)	4 (3.7)	−2.1 (−6.3 to 2.0)	0.44 (0.08 to 2.49)	.35
Catastrophic event	7 (5.4)	6 (5.5)	−0.1 (−5.9 to 5.7)	1.03 (0.33 to 3.22)	.96
Multiorgan failure	16 (12.3)	13 (11.9)	0.4 (−7.9 to 8.7)	1.09 (0.49 to 2.42)	.94
CPC score, No. (%)^d	n = 213	n = 210			.27
0-2	78 (36.6)	88 (41.9)	5.3 (−14.6 to 4.0)	0.80 (0.54 to 1.18)
3-5	135 (63.4)	122 (58.1)		1 [Reference]
Discharge destination in survivors, No. (%)	82	101			.08
Home	52 (63.4)	74 (73.3)	−9.9 (−23.4 to 3.7)	1 [Reference]
Rehabilitation	30 (36.6)	24 (23.8)	12.8 (−0.5 to 26.2)	1.78 (0.93 to 3.38)
Other hospital	0	3 (3.0)

Open in a new tab

Abbreviations: CPC, Cerebral Performance Category; ED, emergency department; ICU, intensive care unit; ROSC, return of spontaneous circulation; Spo₂, oxygen saturation.

^{^a}

For all end points, the difference is the absolute difference.

^{^b}

Logistic regression used for binary outcomes. Multinomial logistic regression used for discharge destination and cause of death. Median regression used for length of stay outcomes.

^{^c}

Seven patients had rearrest in both the prehospital and ED settings.

^{^d}

CPC is a 5-point scale assessing neurologic outcomes after brain damage, with higher scores indicating worse outcomes. A score of 1 or 2 is considered a favorable outcome.

Secondary Outcomes

Of the 9 prespecified secondary outcomes collected up to hospital discharge, 8 were not statistically significantly different between treatment groups. A hypoxic episode prior to intensive care was observed in 31.3% (n = 67) of participants in the intervention group and 16.1% (n = 34) in the standard care group (difference, 15.2% [95% CI, 7.2%-23.1%]; OR, 2.37 [95% CI, 1.49-3.79]; P < .001).

In patients surviving to hospital discharge, 147 of 183 (80.3%) were followed up at 12 months, with 141 consenting to 12-month quality of life questions, 6 deceased, and 19 lost to follow-up (5/82 in the intervention and 14/101 in the standard care group; eTable 4 in Supplement 3). Survival at 12 months was 35% (72/208) in the intervention group and 42% (81/193) in the standard care group. Quality of life scores for the treatment groups are reported in eTable 4 and eFigure 2 in Supplement 3.

Serious Adverse Events

A sustained hypoxic event (Spo₂ <90%) unresponsive to 100% oxygen was seen in 5 participants (2.3%) in the intervention group and in 3 (1.4%) in the standard care group. All 3 patients (1.4%) who had rearrest in the setting of hypoxia (Spo₂ <90%) were in the intervention group.

Subgroup Analyses

Results were consistent across 6 of the 7 prespecified subgroups (Figure 2). The association between treatment and survival varied according to whether the patient received bystander CPR (bystander CPR: OR, 0.48 [95% CI, 0.31-0.75]; no bystander CPR: OR, 1.96 [95% CI, 0.72-5.33]; interaction P = .01). A post hoc examination showed that those in the intervention group who received bystander cardiopulmonary resuscitation had higher rates of hypoxia (Spo₂ < 90%; 30% vs 16%) and rearrest without ROSC (14% vs 4%) than those in the standard care group who received bystander CPR.

Post Hoc Analyses

A post hoc analysis showed no significant difference in survival to discharge when adjusting for known predictors and site (OR, 0.61 [95% CI, 0.36, 1.03]; P = .06). There was no statistical difference between treatment groups for survival to discharge stratified by time from EMS call to randomization (eTable 5 in Supplement 3) or the early withdrawal of life support (<72 hours after randomization) (eTable 6 in Supplement 3).

Discussion

Among patients achieving ROSC after OHCA, targeting an oxygen saturation of 90% to 94%, compared with 98% to 100%, until admission to the ICU did not significantly improve survival to hospital discharge. Although the trial is limited by early termination due to the COVID-19 pandemic, the findings do not support use of an oxygen saturation target of 90% to 94% in the out-of-hospital setting after resuscitation from cardiac arrest.

The outcome in this study contrasts with previous findings of an association between hyperoxia during the early postarrest period and increased neurological injury in animal,³ neonate,²⁵ and adult studies.^4,5 An initial observational study in adult patients with cardiac arrest by Kilgannon et al²⁰ reported that hyperoxia (first Pao₂ in ICU ≥300 mm Hg), compared with normoxia (Pao₂ between 60 and 299 mm Hg), was associated with significantly higher in-hospital mortality (63% vs 45%; difference, 18% [95% CI, 14%-22%]). However, Kilgannon et al²⁰ used a single blood gas measurement. Other smaller observational studies in similar cohorts have found no significant difference in outcomes related to oxygen saturation.^26,27

There have been several preliminary randomized clinical trials comparing titration of oxygen to different levels in the out-of-hospital^8,28,29,30 and in-hospital^31,32 settings in patients after cardiac arrest. An individual patient data meta-analysis of these trials indicated that conservative oxygen therapy was significantly associated with reduced mortality compared with liberal oxygen therapy.⁵ Although this meta-analysis mainly included patients with OHCA (406 of 429), it did not examine differences by the setting of oxygen titration.

The outcomes in the current study suggest possible harm in patients who underwent reduction of oxygen targeting a saturation of 90% to 94% commencing in the out-of-hospital setting. It is possible that the methods of titration used in the current trial resulted in relatively large decreases in the delivery of oxygen, which, in patients with decreased cerebral blood flow due to a low cardiac output, resulted in additional cerebral hypoxic injury. It is also possible that the increase in hypoxic events in the intervention group resulted in myocardial ischemia, which may explain the increased number of rearrests with no ROSC seen in this treatment group.

Titrating oxygen in the out-of-hospital setting is challenging when using the equipment available in this trial. The air-mix settings on the available ventilators did not allow for slow titration and, although titration of oxygen flow into the ventilation bag reservoir is feasible in a laboratory setting,³³ it may be difficult to provide a reliable Fio₂ in practice and this approach may result in increased episodes of oxygen desaturation.³⁰ Therefore, it is suggested that future studies be performed in hemodynamically stable patients after hospital arrival with mechanical ventilators that provide a precise delivery of oxygen.

Additionally, it may be that a target oxygen saturation of 90% to 94% soon after ROSC is too low, noting the recent recommendations from the European Resuscitation Council and European Society of Intensive Care Medicine Guidelines of a target of an oxygen saturation of 94% to 98% during postarrest care.³⁴

Limitations

This study has several limitations. First, the number of patients enrolled was less than the planned enrollment. However, the decision to stop the study was made without the investigators being aware of study patient outcomes and prior to any analysis. Second, the methods of titration of oxygen, to either different oxygen flows to bags with reservoirs or EMS ventilators limited to either 100% oxygen or oxygen with air mix, did not allow for accurate oxygen titration. Third, most trial patients had an arrest of cardiac etiology and the findings may not be applicable to other etiologies. Fourth, as with many out-of-hospital studies, it was difficult to ensure that only eligible patients are enrolled and that the trial protocol was closely followed.

Conclusions

Among patients achieving ROSC after out-of-hospital cardiac arrest, targeting an oxygen saturation of 90% to 94%, compared with 98% to 100%, until admission to the ICU did not significantly improve survival to hospital discharge. Although the trial is limited by early termination due to the COVID-19 pandemic, the findings do not support use of an oxygen saturation target of 90% to 94% in the out-of-hospital setting after resuscitation from cardiac arrest.

Supplement 1.

Trial protocol

Click here for additional data file.^{(1.6MB, pdf)}

Supplement 2.

Statistical analysis plan

Click here for additional data file.^{(302.7KB, pdf)}

Supplement 3.

eAppendix 1. Discrepancies between the Manuscript and the Protocol and Statistical Analysis Plan

eAppendix 2. Prehospital arrest and post-arrest management during the EXACT trial

eTable 1. Reasons for exclusion from the sensitivity analysis

eTable 2. Time from randomization to first arterial blood gas in ED and ICU

eTable 3. Sensitivity analysis of outcomes in eligible cases

eTable 4. Outcomes and quality of life measures at 12-months

eTable 5. An exploratory subgroup analysis of survival to discharge stratified for time from emergence call to randomization

eTable 6. An exploratory analysis of the timing of withdrawal of life-sustaining treatment

eFigure 1. Recorded Mean and Standard Deviation Oxygen Saturation Measurements over the First Six Hours by Treatment Allocation

eFigure 2. Modified Rankin Scores by treatment group at 12 months including in-hospital deaths (excluding those loss-to follow-up)

Click here for additional data file.^{(1.1MB, pdf)}

Supplement 4.

Nonauthor collaborators

Click here for additional data file.^{(110.8KB, pdf)}

Supplement 5.

Data sharing statement

Click here for additional data file.^{(12.1KB, pdf)}

References

1.Kleinman ME, Perkins GD, Bhanji F, et al. ILCOR scientific knowledge gaps and clinical research priorities for cardiopulmonary resuscitation and emergency cardiovascular care: a consensus statement. Resuscitation. 2018;127:132-146. doi: 10.1016/j.resuscitation.2018.03.021 [DOI] [PubMed] [Google Scholar]
2.Kempster K, Howell S, Bernard S, et al. Out-of-hospital cardiac arrest outcomes in emergency departments. Resuscitation. 2021;166:21-30. doi: 10.1016/j.resuscitation.2021.07.003 [DOI] [PubMed] [Google Scholar]
3.Pilcher J, Weatherall M, Shirtcliffe P, Bellomo R, Young P, Beasley R. The effect of hyperoxia following cardiac arrest - a systematic review and meta-analysis of animal trials. Resuscitation. 2012;83(4):417-422. doi: 10.1016/j.resuscitation.2011.12.021 [DOI] [PubMed] [Google Scholar]
4.Holmberg MJ, Nicholson T, Nolan JP, et al. ; Adult Pediatric Advanced Life Support Task Forces at the International Liaison Committee on Resuscitation (ILCOR) . Oxygenation and ventilation targets after cardiac arrest: a systematic review and meta-analysis. Resuscitation. 2020;152:107-115. doi: 10.1016/j.resuscitation.2020.04.031 [DOI] [PubMed] [Google Scholar]
5.Young PJ, Bailey M, Bellomo R, et al. Conservative or liberal oxygen therapy in adults after cardiac arrest: an individual-level patient data meta-analysis of randomised controlled trials. Resuscitation. 2020;157:15-22. doi: 10.1016/j.resuscitation.2020.09.036 [DOI] [PubMed] [Google Scholar]
6.Sandroni C, Cronberg T, Sekhon M. Brain injury after cardiac arrest: pathophysiology, treatment, and prognosis. Intensive Care Med. 2021;47(12):1393-1414. doi: 10.1007/s00134-021-06548-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Stub D, Smith K, Bernard S, et al. ; AVOID Investigators . Air versus oxygen in ST-segment-elevation myocardial infarction. Circulation. 2015;131(24):2143-2150. doi: 10.1161/CIRCULATIONAHA.114.014494 [DOI] [PubMed] [Google Scholar]
8.Bray JE, Hein C, Smith K, et al. ; EXACT Investigators . Oxygen titration after resuscitation from out-of-hospital cardiac arrest: a multi-centre, randomised controlled pilot study (the EXACT pilot trial). Resuscitation. 2018;128:211-215. doi: 10.1016/j.resuscitation.2018.04.019 [DOI] [PubMed] [Google Scholar]
9.Teran F, Abella BS. Oxygen delivery after cardiac arrest: adding fuel to the fire? Resuscitation. 2018;128:A5-A6. doi: 10.1016/j.resuscitation.2018.05.016 [DOI] [PubMed] [Google Scholar]
10.Bray JE, Smith K, Hein C, et al. ; EXACT investigators . The EXACT protocol: a multi-centre, single-blind, randomised, parallel-group, controlled trial to determine whether early oxygen titration improves survival to hospital discharge in adult OHCA patients. Resuscitation. 2019;139:208-213. doi: 10.1016/j.resuscitation.2019.04.023 [DOI] [PubMed] [Google Scholar]
11.The ARC guidelines. Australian Resuscitation Council . Accessed July 20, 2022. https://resus.org.au/the-arc-guidelines
12.Jacobs I, Nadkarni V, Bahr J, et al. ; International Liaison Committee on Resuscitation; American Heart Association; European Resuscitation Council; Australian Resuscitation Council; New Zealand Resuscitation Council; Heart and Stroke Foundation of Canada; InterAmerican Heart Foundation; Resuscitation Councils of Southern Africa; ILCOR Task Force on Cardiac Arrest and Cardiopulmonary Resuscitation Outcomes . Cardiac arrest and cardiopulmonary resuscitation outcome reports: update and simplification of the Utstein templates for resuscitation registries: a statement for healthcare professionals from a task force of the International Liaison Committee on Resuscitation (American Heart Association, European Resuscitation Council, Australian Resuscitation Council, New Zealand Resuscitation Council, Heart and Stroke Foundation of Canada, InterAmerican Heart Foundation, Resuscitation Councils of Southern Africa). Circulation. 2004;110(21):3385-3397. doi: 10.1161/01.CIR.0000147236.85306.15 [DOI] [PubMed] [Google Scholar]
13.Perkins GD, Jacobs IG, Nadkarni VM, et al. ; Utstein Collaborators . Cardiac arrest and cardiopulmonary resuscitation outcome reports: update of the Utstein Resuscitation Registry Templates for Out-of-Hospital Cardiac Arrest: a statement for healthcare professionals from a task force of the International Liaison Committee on Resuscitation (American Heart Association, European Resuscitation Council, Australian and New Zealand Council on Resuscitation, Heart and Stroke Foundation of Canada, InterAmerican Heart Foundation, Resuscitation Council of Southern Africa, Resuscitation Council of Asia); and the American Heart Association Emergency Cardiovascular Care Committee and the Council on Cardiopulmonary, Critical Care, Perioperative and Resuscitation. Circulation. 2015;132(13):1286-1300. doi: 10.1161/CIR.0000000000000144 [DOI] [PubMed] [Google Scholar]
14.Jennett B, Bond M. Assessment of outcome after severe brain damage. Lancet. 1975;1(7905):480-484. doi: 10.1016/S0140-6736(75)92830-5 [DOI] [PubMed] [Google Scholar]
15.Ware J Jr, Kosinski M, Keller SDA. A 12-Item Short-Form Health Survey: construction of scales and preliminary tests of reliability and validity. Med Care. 1996;34(3):220-233. doi: 10.1097/00005650-199603000-00003 [DOI] [PubMed] [Google Scholar]
16.Herdman M, Gudex C, Lloyd A, et al. Development and preliminary testing of the new five-level version of EQ-5D (EQ-5D-5L). Qual Life Res. 2011;20(10):1727-1736. doi: 10.1007/s11136-011-9903-x [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Wilson JT, Pettigrew LE, Teasdale GM. Structured interviews for the Glasgow Outcome Scale and the extended Glasgow Outcome Scale: guidelines for their use. J Neurotrauma. 1998;15(8):573-585. doi: 10.1089/neu.1998.15.573 [DOI] [PubMed] [Google Scholar]
18.Wilson JTL, Hareendran A, Grant M, et al. Improving the assessment of outcomes in stroke: use of a structured interview to assign grades on the modified Rankin Scale. Stroke. 2002;33(9):2243-2246. doi: 10.1161/01.STR.0000027437.22450.BD [DOI] [PubMed] [Google Scholar]
19.Bernard SA, Smith K, Finn J, et al. Induction of therapeutic hypothermia during out-of-hospital cardiac arrest using a rapid infusion of cold saline: the RINSE Trial (Rapid Infusion of Cold Normal Saline). Circulation. 2016;134(11):797-805. doi: 10.1161/CIRCULATIONAHA.116.021989 [DOI] [PubMed] [Google Scholar]
20.Kilgannon JH, Jones AE, Shapiro NI, et al. ; Emergency Medicine Shock Research Network (EMShockNet) Investigators . Association between arterial hyperoxia following resuscitation from cardiac arrest and in-hospital mortality. JAMA. 2010;303(21):2165-2171. doi: 10.1001/jama.2010.707 [DOI] [PubMed] [Google Scholar]
21.Damiani E, Adrario E, Girardis M, et al. Arterial hyperoxia and mortality in critically ill patients: a systematic review and meta-analysis. Crit Care. 2014;18(6):711. doi: 10.1186/s13054-014-0711-x [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Wang CH, Chang WT, Huang CH, et al. The effect of hyperoxia on survival following adult cardiac arrest: a systematic review and meta-analysis of observational studies. Resuscitation. 2014;85(9):1142-1148. doi: 10.1016/j.resuscitation.2014.05.021 [DOI] [PubMed] [Google Scholar]
23.Ball J, Nehme Z, Bernard S, Stub D, Stephenson M, Smith K. Collateral damage: hidden impact of the COVID-19 pandemic on the out-of-hospital cardiac arrest system-of-care. Resuscitation. 2020;156:157-163. doi: 10.1016/j.resuscitation.2020.09.017 [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Orkin AM, Gill PJ, Ghersi D, et al. ; CONSERVE Group . Guidelines for reporting trial protocols and completed trials modified due to the COVID-19 pandemic and other extenuating circumstances: the CONSERVE 2021 Statement. JAMA. 2021;326(3):257-265. doi: 10.1001/jama.2021.9941 [DOI] [PubMed] [Google Scholar]
25.Welsford M, Nishiyama C, Shortt C, et al. ; International Liaison Committee on Resuscitation Neonatal Life Support Task Force . Room air for initiating term newborn resuscitation: a systematic review with meta-analysis. Pediatrics. 2019;143(1):e20181825. doi: 10.1542/peds.2018-1825 [DOI] [PubMed] [Google Scholar]
26.Ihle JF, Bernard S, Bailey MJ, Pilcher DV, Smith K, Scheinkestel CD. Hyperoxia in the intensive care unit and outcome after out-of-hospital ventricular fibrillation cardiac arrest. Crit Care Resusc. 2013;15(3):186-190. [PubMed] [Google Scholar]
27.Vaahersalo J, Bendel S, Reinikainen M, et al. ; FINNRESUSCI Study Group . Arterial blood gas tensions after resuscitation from out-of-hospital cardiac arrest: associations with long-term neurologic outcome. Crit Care Med. 2014;42(6):1463-1470. doi: 10.1097/CCM.0000000000000228 [DOI] [PubMed] [Google Scholar]
28.Kuisma M, Boyd J, Voipio V, Alaspää A, Roine RO, Rosenberg P. Comparison of 30 and the 100% inspired oxygen concentrations during early post-resuscitation period: a randomised controlled pilot study. Resuscitation. 2006;69(2):199-206. doi: 10.1016/j.resuscitation.2005.08.010 [DOI] [PubMed] [Google Scholar]
29.Thomas M, Voss S, Benger J, Kirby K, Nolan JP. Cluster randomised comparison of the effectiveness of 100% oxygen versus titrated oxygen in patients with a sustained return of spontaneous circulation following out of hospital cardiac arrest: a feasibility study: PROXY: post ROSC OXYgenation study. BMC Emerg Med. 2019;19(1):16. doi: 10.1186/s12873-018-0214-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Young P, Bailey M, Bellomo R, et al. Hyperoxic therapy or normoxic therapy after out-of-hospital cardiac arrest (HOT OR NOT): a randomised controlled feasibility trial. Resuscitation. 2014;85(12):1686-1691. doi: 10.1016/j.resuscitation.2014.09.011 [DOI] [PubMed] [Google Scholar]
31.Jakkula P, Reinikainen M, Hästbacka J, et al. ; COMACARE study group . Targeting two different levels of both arterial carbon dioxide and arterial oxygen after cardiac arrest and resuscitation: a randomised pilot trial. Intensive Care Med. 2018;44(12):2112-2121. doi: 10.1007/s00134-018-5453-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
32.Young P, Mackle D, Bellomo R, et al. ; ICU-ROX Investigators and the Australian and New Zealand Intensive Care Society Clinical Trials Group . Conservative oxygen therapy for mechanically ventilated adults with suspected hypoxic ischaemic encephalopathy. Intensive Care Med. 2020;46(12):2411-2422. doi: 10.1007/s00134-020-06196-y [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Young P, Pilcher J, Patel M, et al. Delivery of titrated oxygen via a self-inflating resuscitation bag. Resuscitation. 2013;84(3):391-394. doi: 10.1016/j.resuscitation.2012.08.330 [DOI] [PubMed] [Google Scholar]
34.Nolan JP, Sandroni C, Böttiger BW, et al. European Resuscitation Council and European Society of Intensive Care Medicine guidelines 2021: post-resuscitation care. Intensive Care Med. 2021;47(4):369-421. doi: 10.1007/s00134-021-06368-4 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials