A comparative study of targeted end-tidal carbon dioxide versus standard management to guide resuscitation in emergency department cardiac arrest in Thailand: study protocol for a three-arm randomised controlled trial

Abstract

Background

Cardiac arrest managed in the emergency department (ED) is associated with suboptimal survival outcomes. End-tidal carbon dioxide (EtCO₂) monitoring during cardiopulmonary resuscitation (CPR) reflects cardiac output, tissue perfusion, and pulmonary blood flow. Observational evidence suggests that higher EtCO₂ levels during CPR are associated with return of spontaneous circulation (ROSC), but the effect of actively targeting higher EtCO₂ thresholds remains uncertain. The primary objective of this trial is to determine whether CPR guided by higher EtCO₂ targets improves sustained ROSC compared with a standard EtCO₂ target among patients with ED cardiac arrest.

Methods

This is a single-centre, three-arm, single-blind, superiority randomised controlled trial conducted in the ED of Thammasat University Hospital, Pathum Thani, Thailand. Eligible participants are patients with ED cardiac arrest, including out-of-hospital cardiac arrest patients receiving CPR in the ED and patients with first in-hospital cardiac arrest occurring in the ED. Exclusion criteria include age < 15 years, traumatic cardiac arrest, a do-not-attempt-resuscitation order, and signs of irreversible death. Post-randomisation exclusions include inability to obtain EtCO₂ monitoring through an advanced airway and ROSC before the first EtCO₂ measurement. Consecutive eligible participants will be screened and enrolled by the attending emergency physician. Participants will be randomised in a 1:1:1 ratio using fixed-block randomisation with a block size of six; no stratification variables will be used. The planned sample size is 300 participants, with 100 participants per group. All participants will receive high-quality CPR according to international resuscitation guidelines and will be assigned to one of three EtCO₂ targets: ≥ 10 mmHg, ≥ 20 mmHg, or ≥ 25% of venous partial pressure of carbon dioxide (PvCO₂), with a minimum threshold of 10 mmHg. Peripheral venous PvCO₂ will be obtained using point-of-care testing every 10 min. Chest compression depth and rate will be titrated in 0.5-cm and 10/min increments, respectively, every 30 s until the assigned EtCO₂ target is achieved. The primary outcome is sustained ROSC. Secondary outcomes include survival to hospital discharge and survival with favourable neurological outcome. Safety and harm measures include adverse events related to the study intervention, unanticipated device-related issues, and complications attributed to study-directed chest-compression adjustments. These events will be documented, reported to the coordinating centre and ethics committee as required, and reviewed by the Data Safety Monitoring Board. Analyses will follow both intention-to-treat and per-protocol principles.

Discussion

This trial will evaluate whether higher absolute or patient-specific EtCO₂ targets during CPR improve resuscitation outcomes compared with the standard EtCO₂ threshold in ED cardiac arrest patients.

Trial registration

Thai Clinical Trials Registry (TCTR), TCTR20240619001, registered on 19 June 2024: https://www.thaiclinicaltrials.org/show/TCTR20240619001.

Introduction

Background and rationale {6a}

Cardiac arrest is a critical medical emergency; timely and effective interventions including basic and advanced cardiopulmonary resuscitation (CPR) may improve patient outcomes [1]. Nevertheless, outcomes after CPR remain suboptimal and vary substantially across regions and healthcare systems. In international OHCA literature, survival to hospital discharge is generally reported to be approximately 10% or lower in high-income country settings [2–4]. Regional Asian data from the Pan Asian Resuscitation Outcomes Study showed that survival to hospital discharge varied from 0.5 to 8.5%, and survival with favourable neurological outcome ranged from 1.6 to 3.0% across participating countries. In Thailand, survival to hospital discharge after OHCA was reported to be approximately 4.2%, with favourable neurological outcome in 1.6% of patients, despite comparable return of spontaneous circulation (ROSC) rates with other Asian countries [5].

In the in-hospital setting, a United States study reported that emergency department (ED) cardiac arrest accounted for 30.4% of all in-hospital cardiac arrest cases and was associated with lower survival than inpatient cardiac arrest occurring in other hospital areas, approximately 10% versus 31.6%, respectively [6]. In the local setting, Thammasat University Hospital (TUH) manages approximately 130–150 adult, non-traumatic cardiac arrest cases per year. Previous local data from TUH reported sustained ROSC rates of 53–65% and survival-to-discharge rates of approximately 13.8–14% [7]. These estimates suggest that survival after cardiac arrest remains limited and highlight the need for strategies to improve resuscitation outcomes in the ED.

Survival after cardiac arrest depends on a well-coordinated response that follows the chain of survival and delivers high-quality cardiopulmonary resuscitation (CPR). International guidelines outline clear principles for effective CPR [1, 8]. Rescuers should provide high-quality chest compressions and appropriate ventilation throughout resuscitation. In real-world settings, however, multiple team members perform simultaneous tasks, making direct measurement of CPR quality challenging. Current guidelines therefore recommend devices that provide real-time audiovisual feedback on compression and ventilation quality, together with monitoring of end-tidal carbon dioxide (EtCO₂) to assess CPR effectiveness during resuscitation [1, 8].

EtCO₂ monitoring during CPR is a valuable tool for evaluating CPR effectiveness. Current guidelines recommend measuring EtCO₂ during resuscitation; values ≥ 10 mmHg during CPR are associated with a higher likelihood of ROSC [1, 8]. If EtCO₂ falls below 10 mmHg, rescuers should optimise chest compression quality, assess ventilation adequacy, evaluate compressor fatigue, and rotate providers as needed. Observational studies suggest that higher EtCO₂ during CPR correlates with better outcomes, including increased ROSC and survival [9–12]. However, these data are largely observational and prone to bias with limited control for confounding, so the optimal EtCO₂ target for improving outcomes remains uncertain. EtCO₂ values measured during CPR reflect the cardiac output generated by chest compressions, global organ perfusion, and pulmonary blood flow [8, 13–15]. Higher values likely indicate better circulation during resuscitation.

Objectives {7}

This trial aims to evaluate whether higher EtCO₂-targeted CPR improves clinical outcomes compared with the standard EtCO₂ target of ≥ 10 mmHg in patients with ED cardiac arrest.

The specific aims are:

• 1. To determine whether targeting an EtCO₂ level of ≥ 20 mmHg increases the rate of sustained ROSC compared with the standard EtCO₂ target of ≥ 10 mmHg.

• 2. To determine whether targeting an EtCO₂ level of ≥ 25% of venous partial pressure of carbon dioxide (PvCO₂), with a minimum threshold of 10 mmHg, increases the rate of sustained ROSC compared with the standard EtCO₂ target of ≥ 10 mmHg.

• 3. To evaluate whether higher EtCO₂-targeted CPR improves survival to hospital discharge.

• 4. To evaluate whether higher EtCO₂-targeted CPR improves survival with favourable neurological outcome.

Trial design {8}

This is a single-centre, three-arm, single-blind, superiority randomised controlled trial with a 1:1:1 allocation ratio. The trial is designed to evaluate EtCO₂-targeted CPR strategies in patients with emergency department cardiac arrest (EDCA).

Participants will be randomly assigned to one of three EtCO₂ target groups:

• 1. Standard EtCO₂ target of ≥ 10 mmHg;

• 2. Higher absolute EtCO₂ target of ≥ 20 mmHg; or.

• 3. Patient-specific EtCO₂ target of ≥ 25% of PvCO₂, with a minimum threshold of 10 mmHg.

All participants will receive basic life support, high-quality chest compressions, advanced life support, and post–cardiac arrest care according to international resuscitation guidelines and institutional protocols. The only planned difference between groups is the assigned EtCO₂ target used to guide stepwise adjustment of chest compression depth and rate during CPR [16]. The overall trial design and allocation groups are shown in Fig. 1.

Fig. 1.

Study flow diagram and group allocation for EtCO₂-targeted cardiopulmonary resuscitation. Flowchart showing randomisation of emergency department cardiac arrest patients into three EtCO₂ target groups: ≥ 10 mmHg, ≥ 20 mmHg, and ≥ 25% of PvCO₂ with a minimum of 10 mmHg, with corresponding outcomes of ROSC and survival. EtCO₂: End-tidal Carbon Dioxide; PvCO₂: Venous Partial Pressure of Carbon Dioxide; ROSC: Return of Spontaneous Circulation

Methods: participants, interventions, and outcomes

Study setting {9}

From April 2024 onward, this single-center, three-arm, randomised, single-blind, controlled trial with a 1:1:1 allocation ratio was conducted in the ED of Thammasat University Hospital (TUH) in Pathum Thani, Thailand, by the Research Unit in Emergency Medicine and Emergency Critical Care, Thammasat University. TUH is an 800-bed tertiary academic teaching hospital located in the northern suburbs of Bangkok, serving a population of approximately 1.1 million people. The ED at TUH manages about 60,000 patients annually, including approximately 150 cases of cardiac arrest each year, encompassing both OHCA and IHCA patients. All patients receiving CPR in the ED of TUH are treated under the supervision of emergency physician staff, available 24 h a day. The resuscitation team consists of emergency medicine residents and nurses, all of whom are trained and certified in Advanced Life Support.

Eligibility criteria {10}

Patients with cardiac arrest in the ED at TUH (EDCA) included (1) those with OHCA who either received initial CPR in the ED or experienced a recurrent arrest after arrival and (2) those who had their first IHCA in the ED. For patients with recurrent cardiac arrest after achieving sustained ROSC, only data from the first arrest episode were collected.

Inclusion criteria

All patients who experience cardiac arrest in the ED (EDCA).

Exclusion criteria

• Age < 15 years.

• Cardiac arrest due to trauma.

• Presence of a “do not attempt resuscitation” (DNAR) order.

• Signs of irreversible death (livor mortis, rigour mortis, decapitation, transection, or decomposition).

Post-randomisation exclusion criteria

Inability to perform endotracheal intubation for EtCO₂ measurement.

Sustained ROSC occurs before the first EtCO₂ measurement can be obtained.

Expected participant characteristics: To improve transparency in reporting the expected trial population, the anticipated participant characteristics are summarised in Table 1, following the recommended framework for reporting participant characteristics in trial protocols.

Table 1.

Expected participant characteristics for the trial population

Participant characteristic	People expected to be included in this trial
Age	Patients aged ≥ 15 years who experience ED arrest and meet the trial eligibility criteria will be eligible. Patients aged < 15 years will be excluded. Most participants are expected to be adults and older adults, reflecting the usual age distribution of cardiac arrest patients treated in the ED
Sex	Participants of any biological sex will be eligible. Sex will be recorded as documented in the hospital medical record. No eligibility restriction will be applied on the basis of sex
Gender	Gender identity will not be used as an eligibility criterion. Because this is an emergency cardiac arrest trial using deferred consent, gender identity may not always be available during resuscitation. If documented in the medical record, gender-related information will be recorded where available
Race, ethnicity, and ancestry	No eligibility restriction will be applied on the basis of race, ethnicity, ancestry, or nationality. As the trial is conducted in Pathum Thani, Thailand, most participants are expected to be Thai patients, reflecting the hospital catchment population. Race, ethnicity, ancestry, or nationality will be recorded if available in hospital records
Socioeconomic status	Socioeconomic status will not be used as an eligibility criterion. Direct measures such as income or education may not be routinely available in emergency cardiac arrest records. Where available, proxy information such as health insurance or payment scheme may be recorded from hospital administrative data
Geographic location	Participants will be recruited from the ED of Thammasat University Hospital, Pathum Thani, Thailand. The hospital serves patients from Pathum Thani and surrounding areas in the northern Bangkok metropolitan region
Clinical cardiac arrest population	The trial will include patients with ED non-traumatic cardiac arrest, including OHCA patients receiving cardiopulmonary resuscitation in the ED and patients with first in-hospital cardiac arrest occurring in the ED
Resuscitation eligibility	Participants are expected to be patients for whom active resuscitation is attempted. Patients with traumatic cardiac arrest, a do-not-attempt-resuscitation order, signs of irreversible death, inability to obtain EtCO2 monitoring through an advanced airway, or sustained ROSC before the first EtCO2 measurement will be excluded

Who will take informed consent {26a}

Informed consent will be obtained by the first attending emergency physician. All clinicians involved in patient care were trained in the consent process and in the study protocol before trial initiation. Because cardiac arrest is a life-threatening emergency, written informed consent is deferred under emergency research provisions. Consent will be obtained as soon as feasible after resuscitation. If the patient lacks decision-making capacity, consent will be obtained from a legally authorised representative; if the patient later regains capacity, their own written consent will be obtained before any further data collection.

Additional consent provisions for collection and use of participant data and biological specimens {26b}

No additional biological specimens will be collected beyond those obtained as part of routine clinical care and study procedures. Venous blood samples collected during resuscitation are used for clinical decision-making and study outcomes. No genetic testing will be performed, and any future use of specimens will comply with applicable ethical and regulatory requirements.

Interventions

Explanation for the choice of comparators {6b}

The standard EtCO₂ target of ≥ 10 mmHg is recommended in current international resuscitation guidelines as the minimum threshold associated with a higher likelihood of sustained ROSC. This guideline-based target therefore serves as an appropriate control comparator. The two experimental targets, EtCO₂ ≥ 20 mmHg and EtCO₂ ≥ 25% of PvCO₂ (with a minimum of 10 mmHg), were selected based on observational evidence suggesting that higher EtCO₂ levels during cardiopulmonary resuscitation reflect better cardiac output and perfusion, and are associated with improved sustained ROSC and survival. Comparing these higher targets with the guideline-recommended standard allows assessment of whether more aggressive EtCO₂-guided CPR strategies could enhance resuscitation outcomes in EDCA patients.

Intervention description {11a}

All participants will receive standard resuscitation care according to international cardiac arrest guidelines and institutional protocols. This includes basic life support, high-quality chest compressions, advanced life support, defibrillation when indicated, airway management, intravenous or intraosseous medications, and treatment of reversible causes.

After initial rhythm assessment and the first appropriate resuscitation step, eligible participants will be randomised to one of three EtCO₂-targeted CPR strategies:

• 1. Standard EtCO₂ target group: CPR will be guided by a target EtCO₂ of ≥ 10 mmHg.

• 2. Higher absolute EtCO₂ target group: CPR will be guided by a target EtCO₂ of ≥ 20 mmHg.

• 3. Patient-specific EtCO₂ target group: CPR will be guided by a target EtCO₂ of ≥ 25% of PvCO₂, with a minimum threshold of 10 mmHg.

All groups will receive the same standard resuscitation care. The only planned difference between groups is the assigned EtCO₂ target used to guide adjustment of chest compression depth and rate during CPR. EtCO₂ will be monitored continuously using waveform capnography connected to the defibrillator after placement of an advanced airway. In the patient-specific target group, PvCO₂ will be measured using point-of-care testing from peripheral venous blood obtained during resuscitation and repeated every 10 min during ongoing CPR [17]. The EtCO₂ target in the patient-specific target group will be recalculated after each new PvCO₂ measurement.

All participants will initially receive high-quality CPR, defined as a compression depth of 5 cm, a compression rate of 100/min, full chest recoil, minimisation of interruptions to < 10 s, and appropriate ventilation. Ventilation will be delivered at a ratio of 30:2 before advanced airway placement and at 10 breaths/min after advanced airway placement. Mechanical chest-compression devices capable of setting specific compression depth and rate will be used preferentially to ensure consistent CPR quality throughout resuscitation. If a mechanical device is unavailable, manual chest compressions will be performed using a CPR feedback device to maintain optimal compression parameters.

If the measured EtCO₂ meets or exceeds the assigned target, chest compression parameters will remain unchanged. If EtCO₂ is below the assigned target, chest compression depth will first be increased in 0.5-cm increments up to a maximum depth of 6 cm. If the target remains unmet, the compression rate will then be increased in 10/min increments up to a maximum rate of 120/min. These adjustments will be made every 30 s according to the real-time EtCO₂ value.

If the assigned EtCO₂ target is still not achieved after reaching the maximum permitted compression depth and rate, CPR quality will be reassessed, including hand position, full chest recoil, minimisation of interruptions, chest-compression fraction, ventilation rate, and avoidance of endotracheal drug administration. If the target remains unmet despite these measures, the participant will be classified as having a failed EtCO₂ goal. CPR will then continue according to standard resuscitation guidelines, and the participant will remain in the trial for intention-to-treat analysis.

All participants will be connected to a ZOLL R Series defibrillator equipped with CPR feedback and CPR-Stat-Padz HVP multifunction electrodes, enabling real-time monitoring of compression depth and rate, ventilation rate, waveform capnography, and numerical EtCO₂. CPR metrics, EtCO₂ values, ventilation parameters, and electrocardiographic data will be automatically recorded and transmitted to the hospital database. An example of data recording is shown in Fig. 2, and the detailed workflow is summarised in Figs. 3 and 4.

Fig. 2.

Example of real-time CPR performance and EtCO₂ data recording from defibrillator feedback system. Illustrates automated recording of CPR metrics (compression depth, rate, release velocity, chest-compression fraction) and EtCO₂ trend used for subsequent analysis. The raw EtCO₂ data were continuously recorded throughout the entire duration of CPR. CPR: Cardiopulmonary Resuscitation; EtCO₂: End-tidal Carbon Dioxide

Fig. 3.

EtCO₂-targeted CPR workflow for shockable rhythm (VF/pVT). Algorithm summarising CPR steps, defibrillation sequence, drug administration, and EtCO₂-guided compression adjustments every 30 s. CPR: Cardiopulmonary Resuscitation; EtCO₂: End-tidal Carbon Dioxide. IV: intravenous access; IO: intraosseous access

Fig. 4.

EtCO₂-targeted CPR workflow for non-shockable rhythm (asystole/PEA). Algorithm summarising CPR steps, medication timing, randomisation, and EtCO₂-guided compression adjustments every 30 s. CPR: Cardiopulmonary Resuscitation; EtCO₂: End-tidal Carbon Dioxide. IV: intravenous access; IO: intraosseous access

Criteria for discontinuing or modifying allocated interventions {11b}

Allocated interventions will be discontinued or modified under the following circumstances:

• 1. Return of spontaneous circulation: All EtCO₂-targeted adjustments will stop immediately once ROSC is achieved. Further EtCO₂ values after ROSC will not be used for study-related intervention decisions.

• 2. Failure to meet EtCO₂ target despite maximum CPR parameters: If the EtCO₂ target cannot be achieved after increasing compression depth to 6 cm and rate to 120/min, and after confirming optimal CPR quality (correct hand position, full recoil, minimal interruptions, appropriate ventilation), the participant will be classified as having a “failed EtCO₂ goal.” CPR will continue according to standard guidelines, but no further study-specific adjustments will be made.

• 3. Clinical conditions requiring deviation from allocated target: If treating clinicians determine that continuing study-specific EtCO₂-targeted adjustments would compromise patient safety or interfere with essential medical care (e.g. severe airway difficulties, hemodynamic instability requiring immediate procedures), the intervention may be modified at the clinician’s discretion.

• 4. Inability to obtain EtCO₂ measurements: If endotracheal intubation is unsuccessful or EtCO₂ monitoring cannot be reliably obtained, EtCO₂-targeted adjustments will not be performed; the participant will be managed per standard CPR guidelines.

• 5. Withdrawal of consent: If consent is later declined by a legally authorised representative or by the patient, no further study-related interventions will be performed.

All modifications will be documented, and participants will remain in the trial for intention-to-treat analyses.

Strategies to improve adherence to interventions {11c}

To ensure adherence to the allocated EtCO₂-targeted resuscitation protocol, all clinicians involved in patient care, including emergency physicians, residents, and nurses, were trained in the study procedures before trial initiation. Training included CPR quality standards, use of mechanical chest-compression devices, EtCO₂ monitoring, and the stepwise adjustment of compression depth and rate. Real-time CPR feedback from the defibrillator system (including compression depth, rate, and EtCO₂ values) assists clinicians in maintaining protocol fidelity during resuscitation.

Protocol adherence is monitored through automatic data capture of CPR metrics and EtCO₂ trends by the defibrillator, allowing verification of compression parameters and timing of adjustments. Monthly meetings with the study team reinforce adherence, address protocol deviations, and provide refresher training as needed. Any deviation from the intervention protocol will be documented and reviewed by the study coordinators.

Relevant concomitant care permitted or prohibited during the trial {11d}

All participants will receive standard advanced life support (ALS) and post–cardiac arrest care in accordance with international resuscitation guidelines. All usual treatments for cardiac arrest, such as defibrillation, airway management, intravenous medications, fluid resuscitation, and evaluation and treatment of reversible causes, are permitted during the trial.

No concomitant interventions are prohibited unless they interfere with the study intervention. In particular, no drugs will be administered via the endotracheal tube, as this may affect EtCO₂ measurements. Ventilation will be maintained at guideline-recommended rates to avoid excessive ventilation that could alter EtCO₂ values. All other emergent procedures or treatments deemed necessary by the attending clinicians may be performed at their discretion and will not be restricted by the study protocol.

Provisions for post-trial care {30}

No specific post-trial care is required beyond the standard clinical management routinely provided to patients who experience cardiac arrest. All participants who achieve ROSC will continue to receive post–cardiac arrest care according to institutional protocols and established clinical guidelines. Any complications related to standard clinical care will be managed by the treating team. The study involves no investigational drugs or devices; therefore, no additional compensation or special post-trial arrangements are planned. Participants (or their legally authorised representatives) will be informed that they may contact the study team if concerns arise related to participation in the trial.

Outcomes {12}

The primary outcome is sustained ROSC, defined as return of spontaneous cardiac activity without chest compressions, with evidence of adequate circulation for at least 20 min. Evidence of sustained ROSC includes any of the following: palpable carotid pulse, measurable blood pressure, voluntary patient movement, or the presence of an arterial waveform with an organised rhythm if an arterial line is in place. This outcome will be analysed as a binary final status and reported as the proportion of participants achieving sustained ROSC in each treatment group.

The secondary outcomes are survival to hospital discharge and survival with favourable neurological outcome. Survival to hospital discharge is defined as survival to hospital discharge or being alive in hospital more than 30 days after the ED cardiac arrest. This outcome will be analysed as a binary final status and reported as the proportion of participants surviving in each group.

Survival with favourable neurological outcome is defined as survival to hospital discharge or being alive in hospital more than 30 days with good neurological recovery. Neurological outcome will be assessed using the Cerebral Performance Category (CPC) or Overall Performance Category (OPC) scale. Both scales range from 1 to 5, where 1 indicates good cerebral or overall performance, 2 indicates moderate disability, 3 indicates severe disability, 4 indicates coma or vegetative state, and 5 indicates death. Favourable neurological outcome will be defined as a CPC or OPC score of 1 or 2. This outcome will be analysed as a binary final status and reported as the proportion of participants with favourable neurological survival in each group.

Safety and harm outcomes will include adverse events related to the study intervention, serious adverse events, unanticipated device-related issues, and complications attributed to study-directed chest-compression adjustments. These outcomes will be analysed as binary event occurrence and by event type, and reported as the number and proportion of participants experiencing each event in each treatment group. Safety and harm outcomes will be assessed during resuscitation and throughout the hospital course until discharge. Serious adverse events will be reported to the ethics committee according to institutional requirements and reviewed periodically by the Data Safety Monitoring Board (DSMB).

Serious adverse events will include clinically significant events judged to be possibly, probably, or definitely related to the study intervention, including CPR-related thoracic injury requiring intervention, pneumothorax or haemothorax requiring drainage, major internal organ injury, major bleeding requiring transfusion or procedural treatment, device-related events causing clinically important CPR interruption or delayed defibrillation, protocol-related chest-compression interruptions exceeding 10 s, and serious airway-related complications related to EtCO₂ monitoring. These events will be identified from clinical records, CPR feedback and defibrillator data, imaging results, procedure notes, and investigator reports. Death or failure to achieve ROSC due to the underlying cardiac arrest will be recorded as a clinical outcome and will not be classified as a serious adverse event unless judged to be related to the study intervention.

Participant timeline {13}

Because cardiac arrest occurs suddenly and requires immediate intervention, participant activities follow an event-based timeline rather than scheduled study visits. The timeline includes screening for eligibility at the time of cardiac arrest, randomisation immediately after initial resuscitation steps, application of the allocated EtCO₂-targeted CPR strategy during active resuscitation, and assessment of outcomes during and after the resuscitation period until hospital discharge. The schedule of enrollment, interventions, and assessments is summarised in Fig. 5 (SPIRIT Figure).

Fig. 5.

SPIRIT participant timeline. Schedule of enrollment, randomisation, interventions, and assessments in a three-arm randomised controlled trial of EtCO₂-targeted cardiopulmonary resuscitation (CPR) in emergency department (ED) cardiac arrest patients

Sample size {14}

Survival to hospital discharge is considered the most clinically important outcome in this trial and was used as the basis for the sample size calculation. This outcome reflects downstream patient-centered benefit and has a lower expected event rate than sustained ROSC. Survival to hospital discharge will be analysed as a binary endpoint, defined as survival to hospital discharge or being alive in hospital more than 30 days after ED cardiac arrest. It will not be analysed as a time-to-event outcome in the primary trial analysis.

The sample size was calculated using a two-sample comparison of proportions with equal allocation. Based on previous local data from TUH, the survival-to-discharge rate after ED cardiac arrest was approximately 14% [7]. A survival-to-discharge proportion of 31.6%, reported for inpatient cardiac arrest in other hospital areas, was used as the comparator estimate [6]. Using a two-sided alpha of 0.05, 80% power, and a 1:1:1 allocation ratio, the required sample size was approximately 89 participants per group. To allow for potential attrition, withdrawal of consent, post-randomisation exclusion, or incomplete outcome data, the sample size was increased to 100 participants per group. Therefore, the total planned sample size is 300 participants.

Recruitment {15}

Eligible participants will be recruited consecutively from all cardiac arrest cases presenting to the ED of TUH. Because cardiac arrest occurs unpredictably, no active recruitment strategies are required. All adult patients who experience cardiac arrest in the ED and meet the inclusion criteria will be screened for enrollment by the attending emergency physician.

The high annual volume of EDCA (approximately 150 cases per year) is expected to provide an adequate number of eligible participants to achieve the target sample size within the planned recruitment period. Study investigators will review screening logs regularly to ensure consistent application of eligibility criteria and to monitor recruitment progress.

Assignment of interventions: allocation

Sequence generation {16a}

Participants will be randomly assigned in a 1:1:1 ratio to one of three EtCO₂ target groups using blocked randomisation with a fixed block size of six to maintain balance across groups. The allocation sequence was generated before trial initiation by an independent study coordinator who was not involved in participant enrolment, clinical care, or outcome assessment, using a computer-generated randomisation programme.

Concealment mechanism {16b}

Allocation will be concealed using sequentially numbered, opaque, sealed envelopes (SNOSE) prepared according to the computer-generated randomisation sequence. The envelopes will be assembled by study personnel not involved in patient enrollment or clinical care and stored in a secure, designated area in the ED. Envelopes will be opened in strict numerical order only after confirming that a patient meets all eligibility criteria, with no skipping or replacement permitted.

Implementation {16c}

The computer-generated allocation sequence will be prepared by a study coordinator who is not involved in patient enrollment or direct patient care. Eligible participants will be enrolled consecutively by the first attending emergency physician. After confirming eligibility, a third-year emergency medicine resident or designated study team member will open the next sequentially numbered, opaque, sealed envelope to reveal the treatment assignment. The clinical resuscitation team will then implement the allocated EtCO₂-targeted intervention immediately according to the study protocol.

Assignment of interventions: Blinding

Who will be blinded {17a}

This trial is conducted as a single-blind study. Outcome assessors and data analysts will be blinded to group allocation. Because EtCO₂ targets require real-time adjustments to compression depth and rate, treating clinicians and resuscitation staff cannot be blinded. Participants are unconscious and unable to perceive allocation during cardiac arrest; therefore, participant blinding is inherently maintained.

All data used for outcome assessment (sustained ROSC, survival, neurological outcome) will be coded without indication of treatment group, and analysts will receive only de-identified datasets to preserve blinding.

Procedure for unblinding if needed {17b}

Because treating clinicians are not blinded and only outcome assessors and data analysts are blinded, unblinding is generally not required during the conduct of the trial. No circumstances are anticipated in which unblinding would be necessary for patient safety or clinical management, as all resuscitation is performed according to standard guidelines regardless of allocation.

If unblinding of an outcome assessor or data analyst is required for administrative or regulatory reasons, it will be performed by the principal investigator. The reason for unblinding and the personnel involved will be documented. Unblinding will not affect ongoing participant care.

Data collection and management

Plans for assessment and collection of outcomes {18a}

Baseline information necessary to confirm eligibility and initial rhythm will be assessed before randomisation. Other baseline demographic and clinical variables will be collected after randomisation from resuscitation records, device data, and medical records, but these variables will reflect the participant’s status before or at the time of randomisation.

Outcome data will be collected prospectively using a combination of automated device recording, standard clinical documentation, and medical record review. The primary outcome, sustained ROSC, will be assessed during resuscitation by the attending emergency physician based on predefined clinical criteria, including palpable pulse, measurable blood pressure, spontaneous movement, or arterial waveform activity. The time of ROSC and duration of resuscitation will be recorded in real time.

Secondary outcomes, including survival to hospital discharge and survival with favourable neurological status, will be obtained from hospital medical records. Neurological status during hospitalisation and before discharge will be assessed by a neurologist according to the hospital standard of care. Neurological outcome for the study will be classified using the CPC or OPC scale at hospital discharge or at 30 days if the participant remains hospitalised.

Other trial data will include airway management, defibrillation, medications, treatment of reversible causes, adverse events, protocol deviations, and post–cardiac arrest care. CPR metrics, including compression depth, compression rate, chest-compression fraction, ventilation rate, electrocardiographic rhythm, and EtCO₂ values, will be automatically recorded by the ZOLL R Series defibrillator with integrated CPR feedback and waveform capnography. PvCO₂ and other laboratory variables will be measured using POCT from peripheral venous blood obtained immediately during resuscitation and repeated at 10-min intervals during ongoing CPR. These devices are routinely used in clinical care and maintained according to institutional standards.

Standardised data collection forms will be used to capture demographic characteristics, clinical history, baseline clinical variables, initial rhythm, treatments received, CPR metrics, laboratory results, outcomes, adverse events, and post-cardiac arrest care. Data quality will be promoted through training of data collectors, use of predefined variable definitions, duplicate verification of critical variables, review of automated device data, and routine monitoring by the study coordinators. The standardised data collection form is available from the corresponding author.

Plans to promote participant retention and complete follow-up {18b}

As this study involves patients experiencing cardiac arrest in the ED, participant retention primarily concerns complete capture of in-hospital outcomes rather than scheduled follow-up visits. All enrolled participants will be followed through their entire hospital course until discharge or 30 days, whichever occurs later. Outcome data, including survival and neurological status, will be obtained from electronic medical records, minimising loss to follow-up.

For patients transferred to other facilities, the study team will track outcomes through institutional medical records, transfer summaries, or coordinated communication with receiving hospitals when necessary. Screening logs, outcome checklists, and periodic data audits will be used to ensure completeness of follow-up data. No additional procedures or visits are required from patients or families, reducing the likelihood of missing outcome information.

Data management {19}

Study data will be entered into a secure, password-protected electronic database managed by the Research Unit in Emergency Medicine and Emergency Critical Care. CPR performance metrics from the ZOLL R Series defibrillator will be automatically uploaded to the hospital server, while clinical and outcome data will be extracted from electronic medical records using standardised data collection forms. Data entry will be conducted by trained research staff, with verification of key variables and routine range and consistency checks to ensure data accuracy. All data will be stored on secure institutional servers accessible only to authorised investigators. Identifiable information will be removed and replaced with unique study codes prior to analysis. Data will be backed up regularly in accordance with institutional policies.

Confidentiality {27}

All personal information collected during the study will be kept strictly confidential. Participant data will be recorded using unique study codes, and all identifying information will be removed before analysis. Access to identifiable data will be restricted to authorised study investigators and stored on secure, password-protected institutional servers. Any printed study documents will be kept in locked cabinets within the research unit. No identifiable information will be included in reports, publications, or shared datasets.

Plans for collection, laboratory evaluation and storage of biological specimens for genetic or molecular analysis in this trial/future use {33}

No biological specimens will be collected for genetic or molecular analysis in this trial. Venous blood samples obtained during resuscitation are used solely for routine clinical evaluation and study-related measurements (e.g. PvCO₂, acid–base and electrolyte parameters) and will not undergo genetic testing. These specimens will not be stored for future molecular analysis or biobanking. Any future use of clinically obtained samples will comply with institutional ethical requirements and applicable regulations.

Statistical methods

Statistical methods for primary and secondary outcomes {20a}

Baseline demographic and clinical characteristics will be summarised by treatment group using descriptive statistics. Continuous variables will be reported as mean with standard deviation (SD) if normally distributed, or median with interquartile range (IQR) if non-normally distributed. Categorical variables will be reported as frequencies and percentages. Baseline variables will include age, sex, cardiac arrest type, initial cardiac rhythm, witnessed arrest status, comorbidities, airway management, time intervals, initial EtCO₂, PvCO₂, and other relevant clinical and resuscitation variables.

All statistical analyses will be performed using STATA version 17.0 (StataCorp, College Station, TX, USA). The primary outcome, sustained ROSC, and secondary outcomes, including survival to hospital discharge and survival with favourable neurological status, will be compared among the three EtCO₂ target groups using chi-square tests and risk regression models to estimate effect sizes with 95% confidence intervals. Continuous variables will be compared across groups using analysis of variance (ANOVA) or Kruskal–Wallis tests, as appropriate. No stratification variables will be used in the randomisation; therefore, the primary analysis will not adjust for stratification variables. The final trial results will be reported in accordance with the Consolidated Standards of Reporting Trials (CONSORT) statement, including a CONSORT flow diagram and completed checklist.

Interim analyses {21b}

Four interim analyses will be performed at 8-month intervals to monitor potential treatment-related harm. Stopping guidelines will follow Peto’s method, with significance thresholds set at p = 0.001 for the first three interim analyses and p = 0.05 for the final analysis. Interim analyses will be conducted by an independent DSMB, which will have access to unblinded interim results. The DSMB will make recommendations regarding trial continuation, modification, or early termination, and the Trial Steering Committee will make the final decision.

Methods for additional analyses (e.g. subgroup analyses) {20b}

Pre-specified subgroup analyses will be exploratory and include stratification by initial rhythm (shockable vs. non-shockable), arrest type (OHCA vs. IHCA), age group (≤ 65 vs. > 65 years), and initial EtCO₂ tertiles (if available). Effect modification will be assessed by including interaction terms between treatment group and each subgroup variable in regression models. Subgroup-specific effect estimates with 95% confidence intervals will be reported. No formal adjustments for multiplicity will be applied.

Methods in analysis to handle protocol non-adherence and missing data {20c}

The primary efficacy analysis will follow the intention-to-treat (ITT) principle and will include all randomised participants with observed or imputed outcome data, regardless of protocol adherence or whether the assigned EtCO₂ target was achieved. The primary and secondary outcomes will be analysed according to the originally assigned treatment group.

A per-protocol analysis will also be conducted and will include participants who received the allocated EtCO₂-targeted CPR intervention without major protocol deviations. Participants may be excluded from the per-protocol analysis if the allocated intervention could not be implemented, or if there was a major deviation from the assigned intervention protocol.

Safety analyses will include all participants who received any study-related intervention and will be analysed according to the treatment actually received. Pre-specified subgroup and sensitivity analyses will include participants with available data.

Missing outcome data are expected to be minimal because outcomes will be collected from resuscitation records, automated device data, and hospital medical records. If missing data occur, complete-case analysis will be performed as the primary approach, and multiple imputation may be considered depending on the extent and pattern of missingness. Any exclusions due to missing data, withdrawal of consent, post-randomisation exclusion, or protocol deviation will be documented and reported with reasons.

Plans to give access to the full protocol, participant-level data, and statistical code {31c}

The full study protocol is available upon reasonable request from the corresponding author. De-identified participant-level datasets and statistical code will be made available after publication of the primary results, subject to institutional data-sharing policies and approval by the ethics committee. Any shared data will be anonymised to protect participant confidentiality.

Oversight and monitoring

Composition of the coordinating centre and trial steering committee {5d}

The coordinating centre is the Research Unit in Emergency Medicine and Emergency Critical Care at Thammasat University. It is responsible for day-to-day trial operations, including staff training, participant enrollment, data collection, data verification, protocol adherence monitoring, and communication with the ethics committee. The coordinating centre meets monthly to review recruitment progress, data quality, and any protocol deviations.

The Trial Steering Committee (TSC) provides independent oversight of trial conduct. It comprises senior emergency physicians, a clinical trial methodologist, and a biostatistician who are not involved in daily trial operations. The TSC meets every 6 months, or more frequently if needed, to review trial progress, safety reports, and recommendations from the DSMB. The TSC has the authority to advise on protocol modifications, continuation, or suspension of the study.

Composition of the data monitoring committee, its role and reporting structure {21a}

An independent DSMB will oversee participant safety and review interim analyses. The DSMB consists of experts in emergency medicine, biostatistics, and clinical trial methodology who are independent from the sponsor and have no competing interests. The DSMB reviews unblinded interim data in accordance with the stopping guidelines described in this protocol and provides recommendations to the Trial Steering Committee regarding continuation, modification, or early termination of the trial.

Adverse event reporting and harms {22}

Because the interventions involve adjustments to CPR parameters within guideline-recommended ranges, no investigational drugs or devices are used, and the overall risk is minimal. Any adverse events related to the study intervention will be documented by clinical staff and reported to the coordinating centre. Serious adverse events, including unanticipated device-related issues or complications attributed to study-directed chest compression adjustments, will be reported promptly to the ethics committee in accordance with institutional requirements. All adverse events will be summarised and reviewed periodically by the DSMB.

Frequency and plans for auditing trial conduct {23}

Trial conduct will be audited at least annually by the institutional research oversight office, which is independent from the study investigators and sponsor. Audits will include verification of consent procedures, randomisation integrity, data accuracy, and compliance with the approved protocol. Additional audits may be performed if significant protocol deviations or safety concerns arise.

Plans for communicating important protocol amendments {25}

Any substantial protocol amendments, including changes to eligibility criteria, outcomes, analyses, or study procedures, will be submitted for approval to the Human Research Ethics Committee prior to implementation. Approved amendments will be communicated to all investigators, updated in the trial registry, and included in future publications. If amendments affect participant safety or consent, relevant information will also be communicated to legally authorised representatives or participants who regain capacity.

Dissemination plans {31a}

Results of the trial will be disseminated through peer-reviewed publications, conference presentations, and institutional reports. Summaries of study findings will be made available to participants or their families upon request. The sponsor and investigators do not impose publication restrictions. De-identified datasets and statistical code may be shared after publication in accordance with institutional policies and ethical approvals.

Discussion

This three-arm randomised controlled trial, conducted in the ED of a tertiary medical center, compares three EtCO₂ target strategies during CPR in patients presenting with cardiac arrest. We hypothesise that maintaining higher EtCO₂ targets during ED cardiac arrest, in particular, an EtCO₂ target of ≥ 20 mmHg, will yield higher rates of sustained ROSC than the standard target of ≥ 10 mmHg. This hypothesis is supported by observational evidence linking higher EtCO₂ values with improved circulation and greater likelihood of ROSC [1, 8, 18–24], as higher EtCO₂ during CPR likely reflects better cardiac output and perfusion. In addition, this study seeks to test a patient-specific, relative targeting concept: in some patients with preexisting hypercapnia (elevated PCO₂), an absolute EtCO₂ ≥ 10 mmHg during chest compressions may not accurately reflect true perfusion. We therefore hypothesise that targeting EtCO₂ to at least 25% of the patient’s PvCO₂ will better indicate high-quality chest compressions than applying the same absolute 10-mmHg threshold to all patients.

EtCO₂ refers to the partial pressure of carbon dioxide (PCO₂) measured at the end of exhalation. It reflects the patient’s cardiac output, tissue perfusion, and pulmonary blood flow [1, 8]. Carbon dioxide is produced in metabolically active tissues through aerobic metabolism and is transported via the venous circulation to the heart and lungs, where it is eliminated through exhalation. During chest compressions in cardiac arrest, the cardiac output generated by compressions produces blood flow, allowing EtCO₂ to be detected. Therefore, EtCO₂ can serve as an indicator of chest compression quality and ventilation adequacy. In paediatric studies, increasing compression depth has been shown to correlate with higher EtCO₂ values [25]. Observational studies in adults with both IHCA and OHCA found that for every 10-mm increase in compression depth, EtCO₂ rises by approximately 1.4 mmHg (or about 4%) [11, 26], while an increase in compression rate by 10 compressions per minute increases EtCO₂ by about 1.7% [26]. Moreover, increasing compression depth has a stronger association with higher EtCO₂ levels than increasing compression rate, and excessive ventilation can lead to a decrease in EtCO₂ [16].

This study is grounded in the premise that achieving target EtCO₂ levels requires titrating chest compression depth and rate [16, 25, 26]. Increasing compression depth and rate can improve CPR quality and, in turn, enhance systemic perfusion during cardiac arrest, reflected by higher EtCO₂ values. The findings will highlight the feasibility of real-time, EtCO₂-guided resuscitation adjustments. However, variability in EtCO₂ across intervention groups may occur because physiologic responses to CPR differ among individuals. The study may therefore help delineate patient-level differences and support more personalised, EtCO₂-targeted resuscitation strategies.

EtCO₂ can be measured in both patients without an advanced airway and those with one in place. However, studies have shown that during cardiac arrest, EtCO₂ values measured via bag-mask ventilation are lower than those measured through an advanced airway [23, 27]. The accuracy and reliability of EtCO₂ measurements in non-intubated patients remain limited. Furthermore, earlier endotracheal intubation, within the first 2 min after starting chest compressions, in patients with non-shockable rhythms (the predominant rhythm among ED cardiac arrest cases at TUH) has been associated with a trend toward better survival, although the difference was not statistically significant [28].

In the ED of TUH, there is no available device for EtCO₂ monitoring in patients without an advanced airway. Therefore, in this protocol, all patients will undergo endotracheal intubation to ensure accurate EtCO₂ measurement. Previous studies have confirmed the safety of this practice, and the first-attempt and overall success rates of emergency endotracheal intubation in the TUH ED are very high [29–32]. Patients who achieve sustained ROSC before intubation or in whom advanced airway placement is unsuccessful will be excluded from the study.

Blood sampling to identify the causes of cardiac arrest related to metabolic, acid–base, and electrolyte imbalances is essential for improving survival, as it enables targeted treatment of reversible causes [1, 8]. Point-of-care testing (POCT) provides rapid results at the bedside and can be used to guide management during cardiac arrest [33, 34]. Several studies have demonstrated that POCT improves outcomes in cardiac arrest patients [35–37]. Cardiac arrest patients frequently develop acidemia. Studies in IHCA patients have reported a mean pH of 7.2 and a mean PCO₂ of 59.7 mmHg, with higher PCO₂ values observed in those with more severe acidemia [38, 39]. A linear correlation has also been found between PCO₂ and EtCO₂ in cardiac arrest patients [40]. Elevated PCO₂ may influence EtCO₂ readings; therefore, this study is based on the concept that high-quality chest compressions should generate at least 25% of normal cardiac output, reflected as an EtCO₂ level equal to at least 25% of the patient’s PCO₂. This study will therefore evaluate whether adjusting the EtCO₂ target for each patient to at least 25% of their measured PvCO₂ results in better outcomes than using a fixed standard target.

In this protocol, peripheral venous blood samples will be obtained immediately after CPR initiation for POCT, and follow-up samples will be collected every 10 min during resuscitation. Both intubation and blood sampling procedures must be performed without interrupting chest compressions or delaying defibrillation. In addition, this study will specifically control only the adjustment of CPR metrics during resuscitation to achieve the target EtCO₂ values. All patients will otherwise receive standard care during CPR, including evaluation and treatment of the underlying causes of cardiac arrest, as well as post–cardiac arrest care according to hospital protocols and established clinical guidelines [41–43]. EtCO₂ values measured after sustained ROSC will not be used to guide any treatment related to this study.

Trial status

This trial is being conducted under protocol version 3.0 (March 2024). Participant enrollment began on 26 April 2024 at the ED of TUH. Recruitment will continue until the target sample size is reached, which is expected to be completed by 31 March 2027. The trial is currently in the recruitment phase, with ongoing enrollment of eligible participants.

Abbreviations

CPC

Cerebral Performance Category

CPR

Cardiopulmonary resuscitation

DNAR

Do not attempt resuscitation

DSMB

Data Safety Monitoring Board

ED

Emergency department

EDCA

Emergency department cardiac arrest

EtCO₂

End-tidal carbon dioxide

IHCA

In-hospital cardiac arrest

IPCA

Inpatient cardiac arrest

IQR

Interquartile range

OHCA

Out-of-hospital cardiac arrest

OPC

Overall Performance Category

pH

Potential of hydrogen

pVT

Pulseless ventricular tachycardia

PEA

Pulseless electrical activity

POCT

Point-of-care testing

PvCO₂

Venous partial pressure of carbon dioxide

ROSC

Return of spontaneous circulation

SD

Standard deviation

IQR

Interquartile range

ANOVA

Analysis of variance

SNOSE

Sequentially Numbered, Opaque, Sealed Envelopes

TSC

Trial Steering Committee

VF

Ventricular fibrillation

Authors’ contributions {31b}

W.S., and P.T. had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Concept and design: W.S., C.L., K.K., I.I., K.D., N.D., I.D., Y.S., T.U., C.P., and N.C. Acquisition, analysis, or interpretation of data: W.S., and P.T. Drafting of the manuscript: W.S., C.L., K.K., I.I., K.D., N.D., I.D., Y.S., T.U., C.P., and N.C. Critical review of the manuscript for important intellectual content: W.S., C.L., K.K., I.I., K.D., N.D., I.D., Y.S., T.U., C.P., and N.C. Administrative, technical, or material support: W.S. Supervision: W.S., C.L., K.K., I.I., K.D., N.D., I.D., Y.S., T.U., C.P., and N.C. All authors approved the final version of the manuscript for publication.

Funding {4}

This study was supported by the Research Fund of the Faculty of Medicine, Thammasat University (Fiscal Year 2024), and by the Thailand Science Research and Innovation (TSRI) Fundamental Fund (Fiscal Year 2025, project code 206530).

Data availability {29}

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate {24}

The trial was approved by the Human Research Ethics Committee, Faculty of Medicine, Thammasat University (MTU-EC-EM-4–026/67). Given the sudden, life-threatening nature of cardiac arrest, written informed consent was deferred under emergency research provisions. Written informed consent was sought as soon as practicable after resuscitation. If the patient lacked capacity, consent was obtained from a legally authorised representative; if the patient subsequently regained capacity, their own consent was obtained before any further data collection.

Consent for publication {32}

Not applicable.

Competing interests {28}

The authors declare no competing interests.