Value of End-tidal CO2 measurement during pediatric cardiac arrest (PLS #827): Scoping Review

profile avatar

ILCOR staff

To read and leave comments, please scroll to the bottom of this page.

Conflict of Interest Declaration

The ILCOR Continuous Evidence Evaluation process is guided by a rigorous ILCOR Conflict of Interest policy. The following Task Force members and other authors were recused from the discussion as they declared a conflict of interest: none applicable

The following Task Force members and other authors declared an intellectual conflict of interest and this was acknowledged and managed by the Task Force Chairs and Conflict of Interest committees: none applicable

Task Force Scoping Review Citation

Van de Voorde P, Atkins DL , Maconochie I, Aicken R, Bingham R, Couto TB, de Caen A, Guerguerian AM, Nadkarni V, Ng KC, Nuthall G, Ong G, Reis A ,Schexynader S, Tijssen J , Scholefield B, Scholefield B on behalf of the International Liaison Committee on Resuscitation Paediatric Life Support Task Force. on behalf of the International Liaison Committee on Resuscitation Paediatric Life Support Task Force. Value of End-tidal CO2 measurement during pediatric cardiac arrest. Scoping Review and Task Force Insights [Internet] Brussels, Belgium: International Liaison Committee on Resuscitation (ILCOR) Pediatric Life Support Task Force, 2020 Jan 09. Available from:

Methodological Preamble and Link to Published Scoping Review

The continuous evidence evaluation process started with a scoping review of EtCO2 and EtCO2-directed resuscitation in paediatric cardiac arrest, conducted by the ILCOR PLS Task Force Scoping Review team. Evidence from adult and pediatric literature was sought and considered by the PLS Task Force.

Scoping Review



The PICOST (Population, Intervention, Comparator, Outcome, Study Designs and Timeframe)

Population: Infants & Children in any setting (in-hospital or out-of-hospital) with cardiac arrest

Intervention: the presence of variables -images, cut-off values or trends- during CPR (intra-arrest) that can provide physiologic feedback to guide resuscitation efforts, namely: end-tidal CO2

Comparators: the absence of such factors -images, cut-off values or trends.

Outcomes: Any clinical outcome.

Study Designs: STEP 1: Randomized controlled trials (RCTs) and non-randomized studies (non-randomized controlled trials, interrupted time series, controlled before-and-after studies, cohort studies) that concern directly the population and intervention described above are eligible for inclusion. If it is anticipated that there will be insufficient studies from which to draw a conclusion, case series may be included in the initial search. The minimum number of cases for a case series to be included can be set by the ESR after discussion with the priority team or taskforce. Unpublished studies (e.g., conference abstracts, trial protocols) are excluded.
STEP 2: the same study designs and/or existing systematic or scoping reviews not directly concerning the population or intervention defined above but considered informative as additional evidence – taking into account severe indirectness- for the development of the final taskforce insights.

Timeframe: For STEP 1, all languages are included, as long as there is an English abstract. We searched articles from 2015 onwards. For STEP 2, if a systematic or scoping review of high quality (as per AMSTAR 2 tool) is identified, search can be limited to beyond data and/or scope of that review.

Active and Reposed PICOs Related to scope of work for this PICOST:

P 827 EtCO2 Monitoring During CPR

2015 Consensus on Science:

We did not identify any evidence to address the important outcome of survival to hospital discharge or the critical outcome of neurologically intact survival. For the important outcome of ROSC, we identified very-low-quality evidence (downgraded for very serious indirectness and imprecision) from 1 pediatric animal RCT study that showed EtCO2-guided chest compressions are as effective as standard chest compressions optimized by marker, video, and verbal feedback.{Hamrick 2014 e000450)

2015 Treatment Recommendation:

The confidence in effect estimates is so low that the panel decided a recommendation was too speculative.

Search Strategies

We searched PUBMED and Embase with consideration of the predefined inclusion criteria (initial: 15/08/2020 - final review: 10/01/2020). We combined the following terms using Bolean operators: life support care, cardiopulmonary resuscitation, ROSC, heart arrest, cardiac arrest using both individual (ti,ab,kw) and related MESH terms, as well as exploded terms within Embase. We combined these with the terms: end tidal carbon dioxide, carbon dioxide end tidal, end tidal pCO2, EtCO2, capnography (again both individual and MESH terms, and Embase exploded terms). Complete search strategies used are included at the end of this document.

We identified 429 articles after elimination of duplicates. For the subsequent screening by title we excluded only those studies that clearly did not have EtCO2 as study focus or obviously had one of the other pre-defined exclusion criteria.

We included 123 abstracts for review. 6 were deleted because of being duplicates. Two reviewers then evaluated these 117 abstracts for final inclusion. 100% consensus was reached after a first blinded individual review ( and a subsequent discussion about 1 abstract (2 abstracts considered for step 1, 42 for step 2). After subsequent full text review, we finally identified 2 observational trials for step 1. In step 2 we identified 3 adult studies and 6 animal studies (from 2015 onwards). Finally, a follow-up review (10/01/2020) identified 4 additional adult papers.

Inclusion and Exclusion criteria

As part of a cluster of scoping reviews focussing on those variables that can be measured or monitored during the provision of ALS and before ROSC. The variables under consideration are those that have potential to provide (physiologic and/or haemodynamic) feedback to the resuscitating team, potentially guiding their actions to improve outcome, and which might predict outcome. Inclusion criteria as per PICOST above.

Exclusions include:

1/ Newborn at Delivery

2/ Pre-arrest features, not influenced by ALS: time of day, location, bystander CPR, gasping, age, etiology, initial rhythm, unwitnessed…

3/ Standard ALS interventions e.g. ventilation strategies, fluids, firm surface, medications given, eCPR, length and Quality of CPR….

4/ Post-ROSC parameters such as lactate clearance, post-arrest rhythm, hypotension nor any actions to provide neuroprotective care post-ROSC.

Adult data are only considered in step 2 as both the etiology and pathophysiology of paediatric arrest differs substantially (serious indirectness). Moreover, the complexity or impact of obtaining the variable under study will likely differ e.g. advanced airway for capnography. Finally, animal data are considered in step 2 but -as human data are available- mostly to inform the knowledge gaps and possible future research agenda.

Outcome might include: Quality of CPR, ROSC, survival to discharge, (changes in) functional outcome at discharge or other time points.

Data Tables

Non-RCT OR OBSERVATIONAL (acronym); Author; Year Published

Study type/design; Study size (N)

Patient Population
(inclusion criteria)

1. Stine 2019 e01871

Retrospective single-centre observational study; N=49

IHCA (PICU-CVICU) infants 6m;
US, 2008-2012

Results primary endpoints
(P value; OR or RR; & 95% CI)

Ability of EtCO2 to predict ROSC: The highest positive predictive values were seen for EtCO2 values between 17 and 18 mmHg, reaching a PPV of 0.885

Summary/Conclusion - Comments

The authors point out the importance of this as an alternative to the still existing practice of evaluation by auscultation (which may cause a long hands-off time). However, auscultation is (or should be) clearly only part of practice in infants at delivery (transition at birth) and not in infant CPR anyhow.

2. Berg 2018 173

Prospective. multicentre observational study; N= 43 with 48 CPR events

IHCA (60% <1y old)
US, 2013-2016

Results primary endpoints
(P value; OR or RR; & 95% CI)

No association was found between any mean EtCO2 (as such or per CPR minute epoch) and any outcome (ROSC, survival…).

Summary/Conclusion - Comments

In general, the CA events described where of short duration (mean 5 minutes, 65% less than 10 minutes) and the quality of CPR was presumed to be high (dedicated teams, mostly on ICU). This resulted in a ROSC fraction of 73% (and a survival to discharge in 37.2%). Importantly Berg et al found no association between any mean EtCO2 (as such or per CPR minute epoch) and any outcome. From the three patients with EtCO2 below 10 mmHg for every minute of epoch, there was one survivor. There was a relation of EtCO2 with ventilation rate in that the mean EtCO2 decreased 3.6 mmHg [1.3-6] for every 10/minute increase in ventilation. Surprisingly, the mean ventilation rate for these children in CA was 29/minute [24-35] (being mechanically ventilated). On the other hand, there was no correlation of EtCO2 with diastolic blood pressure targets, which drives coronary perfusion. Three patients who reached proper diastolic targets, did so despite a mean etC02 below 15 mmHg. In two it was < 10 mmHg.

Task Force Insights

1. Why this topic was reviewed.

Ideally, physiologic monitoring and feedback to the clinician during cardiac arrest resuscitation would allow rescuers to monitor (and adjust) quality of CPR, and to predict (and influence) the likelihood of return of spontaneous circulation and subsequent neurologic recovery. As such, this physiologic monitoring could lead to a form of ‘individualised’ CPR, where actions are altered to match with individual needs and responses of the victim in cardiac arrest.

End-tidal carbon dioxide (EtCO2) monitoring was initially recommended to confirm tracheal tube placement, but also seemed to offer an estimate of chest compression effectiveness and indirectly of cardiac output, pulmonary blood flow, and coronary perfusion pressure. A rapid increase in EtCO2 may be associated with ROSC, and sustained decline or persistently low values may be associated with the absence of ROSC.

2. Narrative summary of evidence identified

We identified two observational studies {Berg 2018 173; Stine 2019 e01871}. Only the study of Berg et al seems to really inform our PICOST question. Berg et al found no association between any mean EtCO2 (as such or per CPR minute epoch) and any outcome. From the three patients with EtCO2 below 10 mmHg for every minute of epoch, there was one survivor. There was a relation of EtCO2 with ventilation rate in that the mean EtCO2 decreased 3.6 mmHg [1.3-6] for every 10/minute increase in ventilation. Surprisingly, the mean ventilation rate for these children in CA was 29/minute [24-35] (being mechanically ventilated). On the other hand, there was no correlation of EtCO2 with diastolic blood pressure targets, which drives coronary perfusion. Three patients who reached proper diastolic targets, did so despite a mean EtC02 below 15 mmHg. In two it was < 10 mmHg.

To further explore the topic, we also looked at indirect evidence from adult and animal studies (STEP 2).

As part of the ILCOR 2015 evidence evaluation process a systematic review (and meta-analysis) of the use of EtCO2 in adult cardiac arrest was performed. A detailed systematic review was published in 2018 {Paiva 2018 }. This systematic review included 17 observational studies (n=6198; studies up till December 2016), of which five were appropriate for meta-analysis. The majority concerned OHCA. There were no paediatric studies found. They found only limited evidence (of at most low certainty) to suggest a relation of EtCO2 above 10 mmHg and 20 mmHg respectively with increased likelihood for ROSC. Values of EtCO2 below 10 mmHg after 20 minutes of CPR had a 0.5% likelihood of ROSC, and this might inform - according to the authors - discussions about futility.

Further on, Sheak et al {Sheak 2015 149} described a multicentre cohort of 583 adult patients (constituting 29028 fifteen second epochs). For every 10 mm compression depth increase, EtCO2 increased 1.4 mmHg. For every increase of ventilation rate with 10 breaths/minute, EtCO2 decreased 3mmHg.

Sutton et al {Sutton 2016 76} described 803 adult OHCA (as part of a much larger cohort looking at physiologic variables to monitor CPR quality). Survival to discharge and favourable neurological outcome were significantly higher if the EtCO2 during CPR was above 10 mmHg. The Odds Ratio for survival to discharge was 2.41 [95% CI 1.35-4.3].

Savastano et al {Savastano 2017 71} described 207 shock events in 62 patients with VF-cardiac arrest. Shock success was clearly different for those events with ‘mean EtCO2 in the minute before shock’ values below 20mmHg (50%) versus above 30mmHg (78%). None of the shocks with EtCO2 values below 7 while all of the shocks with EtCO2 >45 were successful.

Poppe et al (Poppe 2019 524) reported data from adult non-shockable OHCA victims (with an advanced airway in situ; 2013-2015, Vienna). Only 526 from a total group of 2223 patients met the inclusion criteria. They found that the odds of sustained ROSC and survival were significantly higher for patients presenting with higher values of initial EtCO2 (>45 mmHg): 3.59 [95% CI 2.19 to 5.85] P = 0.001 and 5.02 [95% CI 2.25 to 11.23] P = 0.001, respectively. On the contrary EtCO2 levels less than 20 mmHg were associated with significantly poorer outcomes (OR sustained ROSC 0.29 [95% CI 0.17; 0.49]); there was however no significant difference in survival between the EtCO2 groups 20 to 45 mmHg and less than 20 mmHg.

Skulec et al (Skulec 2019 334) observed 30 patients (results available in 18) with pre-hospital CA comparing EtCO2 levels with results from transthoracic echocardiography. The compression index of LV (LVCI) and RV (RVCI) was calculated as (maximal - minimal/maximal diameter) × 100. Maximal compression index (CImax) defined as the value of LVCI or RVCI, whichever was greater was also assessed. They identified a positive correlation of EtCO2 with LVCI (r = 0.672, p < 0.001) and RVCI (r = 0.778, p < 0.001). The strongest correlation was between EtCO2 and CImax (r = 0.859, p < 0.001). We identified that a CImax cut-off level of 17.35% predicted to reach an EtCO2 level > 20 mmHg with 100% sensitivity and specificity.

Javaudin et al (Javaudin 2019 1) reported data from the French OHCA registry (retrospective, multicentre, 2011-2018). EtCO2 was available in 32249 of 53048 eligible adults. They specifically looked at the maximum value of EtCO2 during CPR. The AUROC of that maximum value to achieve ROSC was 0.887 [95% CI 0.875-0.898] in traumatic OHCA, 0.772 [95% CI 0.765-0.780] in suspected cardiac etiology and 0.802 [95% CI 0.791-0.812] in suspected respiratory etiology. The threshold with no survivors at d-30 was <10 mmHg for traumatic etiologies and <6 mmHg for suspected cardiac and respiratory causes. The probability of ROSC increased with the value of EtCO2 in all studied etiologies.

Finally, Engel et al (Engel 2019 174) performed a prospective observational study in 176 CA adults resuscitated in the ED, comparing EtCO2 and Cerebral Oximetry (CerOx) for ROSC prediction. AUC predictors of ROSC were: last 5 min trend [CerOx = 0.82 ; EtCO2 = 0.74], delta first to last [CerOx = 0.86 ; EtCO2 = 0.73], the penultimate minute [CerOx = 0.81 ; EtCO2 = 0.76], and final minute [CerOx = 0.89 ; EtCO2 = 0.77]

Finally, we also reviewed recent animal research date (from 2015 onwards). The identified data came from three research groups.

Lampe et al {Lampe 2015 69} found in a swine model of VF arrest no significant correlation between EtCO2 and compression depth or blood flow.

Hamrick et al {Hamrick 2017 e575}, in a piglet-model of cardiac arrest, compered real time video/verbal feedback with EtCO2-directed CPR. In the latter group there were 7 survivors (out of 14 piglets), compared to only 2 in the verbal feedback group. There was no effect of epinephrine observed. Interestingly, compression rates largely exceeding current guidelines (143 ± 10/minute) were needed in the EtCO2 directed group.

O’Brien et al {O’Brien 2018 e009728} in 10 piglets also identified that a faster compression rate was necessary to meet EtCO2 goals but saw no changes in myocardial perfusion pressure when CPR was EtCO2-directed.

Hamrick et al {Hamrick 2019 e352} in an asphyxia arrest piglet model found a clear difference in ROSC rate for EtCO2-directed resuscitation (9 vs. 3 survivors out of 14) for an asphyxia duration of 17 minutes. This difference was lost if the asphyxia time was prolonged to 23 minutes.

Morgan et al {Morgan 2016 6} did an RCT in 60 3-month-old swine, with primary VF or asphyxia-associated arrests. The Area-under-curve (AUC) for diastolic blood pressure was clearly superior to EtCO2 for survival (0.82 vs 0.6).

Ryu et al {Ryu 2016 1012} found in 16 piglets a clear relation between compression depth and both systolic blood pressure (immediate effect, AUC 0.895-0.939) and EtCO2 (similar AUC but delay of about 30 seconds). Epinephrine administration changed blood pressure but not EtCO2 values.

3. Narrative Reporting of the task force discussions
Cardiac arrest, and especially OHCA, still has an overall poor prognosis. High quality CPR improves outcome, but what constitutes the best possible CPR for an individual patient is still based on limited evidence and probably differs between patients and etiologies. Having better parameters to guide CPR and adjust it to the need of the individual patient is therefore essential.

Importantly, an advanced airway is required to accurately monitor EtCO2. In children, as in adults, the desirable effects of placing an advanced airway during CPR needs to be balanced by the potential undesirable effects (for this we refer to the specific COSTR on advanced airway placement in paediatric cardiac arrest). Both identified paediatric studies only focused on patients in the intensive care unit, many of whom are already intubated before arrest, which might be considered a specific subpopulation.

To further inform our insights, we also reported on adult and animal studies, however pathophysiological differences between adults and children are such that extrapolation should be done very cautionary (serious indirectness presumed).

EtCO2 is thought to relate to cardiac output and perfusion. However, it was not associated with diastolic blood pressure nor with any pre-defined outcomes in the Berg et al study{Berg 2018 173}. This might be because EtCO2 is also affected by minute volume and ventilation:perfusion matching. It is important to recognise that the study by Berg et al was only descriptive in nature and thus at no point evaluated of the outcomes associated with EtCO2-directed CPR (regardless of current guidelines).

This scoping review has not identified sufficient new evidence to prompt either a new systematic review or reconsideration of current resuscitation guidelines/treatment recommendations.

The use of EtCO2 to confirm endotracheal tube positioning is beyond the scope of the current review.

Knowledge Gaps

The true value of EtCO2-directed resuscitation stills needs to be clarified in future research. Animal data suggest EtCO2-directed resuscitation might make a difference in outcome but only if deviation from existing guidelines is allowed at the same time. Importantly its place will also have to be evaluated in relation to other intra-arrest factors like arterial diastolic blood pressure.

Further on, the interpretation and use of EtCO2 might be different for different cardiac arrest circumstances and aetiologies (eg. asphyxia-related versus primary VF, initial rhythm) and this too should be focus of future research.


  1. Berg RA, Reeder RW, Meert KL, Yates AR, Berger JT, Newth CJ, Carcillo JA, et al; Eunice Kennedy Shriver National Institute of Child Health and Human Development Collaborative Pediatric Critical Care Research Network (CPCCRN) Pediatric Intensive Care Quality of Cardio-Pulmonary Resuscitation (PICqCPR) investigators. End-tidal carbon dioxide during pediatric in-hospital cardiopulmonary resuscitation. Resuscitation. 2018; 133:173-17
  2. Engel TW 2nd, Thomas C, Medado P, Bastani A, Reed B, Millis S, O'Neil BJ. End tidal CO(2) and cerebral oximetry for the prediction of return of spontaneous circulation during cardiopulmonary resuscitation. Resuscitation. 2019; 139:174-181.
  3. Hamrick JL, Hamrick JT, O'Brien CE, Reyes M, Santos PT, Heitmiller SE, Kulikowicz E, et al. The Effect of Asphyxia Arrest Duration on a Pediatric End-Tidal CO2-Guided Chest Compression Delivery Model. Pediatr Crit Care Med. 2019; 20(7):e352-e361.
  4. Hamrick JT, Hamrick JL, Bhalala U, Armstrong JS, Lee JH, Kulikowicz E, Lee JK, et al. End-Tidal CO2-Guided Chest Compression Delivery Improves Survival in a Neonatal Asphyxial Cardiac Arrest Model. Pediatr Crit Care Med. 2017; 18(11):e575-e584.
  5. Javaudin F, Her S, Le Bastard Q, De Carvalho H, Le Conte P, Baert V, Hubert H, Montassier E, Lascarrou JB, Leclère B; GR-RéAC. Maximum Value of End-Tidal Carbon Dioxide Concentrations during Resuscitation as an Indicator of Return of Spontaneous Circulation in out-of-Hospital Cardiac Arrest. Prehosp Emerg Care 2019; 31:1-7
  6. Lampe JW, Tai Y, Bratinov G, Shoap W, Lin Y, Kaufman CL, Becker LB. End tidal carbon dioxide is not a good predictor of compression depth or blood flow during CPR. Resuscitation 2015; 96 Suppl1:69
  7. Morgan RW, French B, Kilbaugh TJ, Naim MY, Wolfe H, Bratinov G, Shoap W, Hsieh TC, Nadkarni VM, Berg RA, Sutton RM. A quantitative comparison of physiologic indicators of cardiopulmonary resuscitation quality: Diastolic blood pressure versus end-tidal carbon dioxide. Resuscitation. 2016; 104:6-11
  8. O'Brien CE, Reyes M, Santos PT, Heitmiller SE, Kulikowicz E, Kudchadkar SR, Lee JK, et al. Pilot Study to Compare the Use of End-Tidal Carbon Dioxide-Guided and Diastolic Blood Pressure-Guided Chest Compression Delivery in a Swine Model of Neonatal Asphyxial Cardiac Arrest. J Am Heart Assoc. 2018; 7(19):e009728
  9. Paiva EF, Paxton JH, O'Neil BJ. The use of end-tidal carbon dioxide (ETCO(2)) measurement to guide management of cardiac arrest: A systematic review. Resuscitation. 2018; 123:1-7.
  10. Poppe M, Stratil P, Clodi C, Schriefl C, Nürnberger A, Magnet I, Warenits AM, Hubner P, Lobmeyr E, Schober A, Zajicek A, Testori C. Initial end-tidal carbon dioxide as a predictive factor for return of spontaneous circulation in nonshockable out-of-hospital cardiac arrest patients: A retrospective observational study. Eur J Anaesthesiol. 2019; 36(7):524-530
  11. Ryu SJ, Lee SJ, Park CH, Lee SM, Lee DH, Cho YS, Jung YH, Lee BK, Jeung KW. Arterial pressure, end-tidal carbon dioxide, and central venous oxygen saturation in reflecting compression depth. Acta Anaesthesiol Scand. 2016; 60(7):1012-3
  12. Savastano S, Baldi E, Raimondi M, Palo A, Belliato M, Cacciatore E, Corazza V, et al. End-tidal carbon dioxide and defibrillation success in out-of-hospital cardiac arrest. Resuscitation 2017; 121: 71-75
  13. Sheak KR, Wiebe DJ, Leary M, Babaeizadeh S, Yuen TC, Zive D, Owens PC, et al. Quantitative relationship between end-tidal carbon dioxide and CPR quality during both in-hospital and out-of-hospital cardiac arrest. Resuscitation. 2015; 89:149-54.
  14. Skulec R, Vojtisek P, Cerny V. Correlation between end-tidal carbon dioxide and the degree of compression of heart cavities measured by transthoracic echocardiography during cardiopulmonary resuscitation for out-of-hospital cardiac
  15. arrest. Crit Care. 2019;23(1):334. Stine CN, Koch J, Brown LS, Chalak L, Kapadia V, Wyckoff MH. Quantitative end-tidal CO(2) can predict increase in heart rate during infant cardiopulmonary resuscitation. Heliyon. 2019; 5(6): e01871
  16. Sutton RM, French B, Meaney PA, Topjian AA, Parshuram CS, Edelson DP, Schexnayder S, et al; American Heart Association’s Get With The Guidelines–Resuscitation Investigators. Physiologic monitoring of CPR quality during adult cardiac arrest: A propensity-matched cohort study. Resuscitation. 2016; 106:76-82

Scoping Review


Add new comment

Please indicate conflict of interest

Something went wrong. Please try again in a few moments. If the problem persists, please contact your administrator.

Add comment as       or   

Sort by

Time range




Review Type