Advanced Airway Interventions in Pediatric Cardiac Arrest
Citation
Nuthall G, Van de Voorde P, Atkins DL, Aickin RP, Bingham R, Couto TB, de Caen AR, Guerguerian A-M, Meaney PA, Nadkarni VM, Ng KC, Nuthall GA, Ong GYK, Reis AG, Schexnayder SM, Shimizu NS, Tijssen JA, Lavonas EJ, Ohshimo S, Nation K, Nolan J, Morrsion L, Maconochie IK. Advanced Airway Interventions in Pediatric Cardiac Arrest- Paediatric Consensus on Science with Treatment Recommendations [Internet] Brussels, Belgium: International Liaison Committee on Resuscitation (ILCOR) Paediatric Advanced Life Support Task Force, 2019 January xxxxx. Available from: http://ilcor.org
Methodological Preamble and Link to Published Systematic Review
The continuous evidence process for the production of Consensus of Science and Treatment Recommendations (CoSTR) started with a systematic review of advanced airway interventions (tracheal intubation and supraglottic airway placement) compared with bag mask ventilation only (BMV) in the management of infants and children in cardiac arrest.[Lavonas 2018; PROSPERO CRD 42018102430 citation] The search was conducted by the Knowledge Synthesis Unit at St. Michael’s Hospital, Toronto, Ontario, Canada, with involvement of clinical content experts. Evidence was sought and considered by the Pediatric Life Support Task Force.
To insert: SR citation with link
PICOST (Population, Intervention, Control, Outcomes, Study design, and Timeframe)
Population: Infants and children in any setting (in-hospital or out-of-hospital) who have received chest compressions or a defibrillation dose on whom CPR is being performed.
Intervention: Placement of an advanced airway device.
Comparators: Primary: Bag-mask ventilation, alone or with non-advanced airway interventions., Secondary: another advanced airway device
Outcomes: Any clinical outcome.
Study Designs: Randomized controlled trials (RCTs) and non-randomized studies (non-randomized controlled trials, interrupted time series, controlled before-and-after studies, cohort studies) of pediatric patients are eligible for inclusion. If there are insufficient studies from which to draw a conclusion, case series of 4 or more may be included. Case reports, unpublished studies and non-human studies are excluded. Timeframe and Languages: All years and all languages are included as long as there is an English abstract. Unpublished studies (e.g., conference abstracts, trial protocols) are excluded. The last search was performed on September 24, 2018
Studies comparing tracheal intubation (TI) with bag mask ventilation alone (BMV):
Fourteen studies were included in the systematic review comparing TI with BMV. Only 1 study provided clinical trial data. (Gausche 2000 783) Three studies provided propensity-adjusted cohort data. (Andersen 2016 1786) (Hansen 2017 51) (Ohashi-Fukuda 2017 66) Eight studies provided retrospective cohort data amenable to meta-analysis.(Abe 2012 612) (Aikian 1989 489) (Deasy 2010 1095) (del Castillo 2015 340) (Guay 2004 373) (Pitetti 2002 283) (Sirbaugh 1999 174) (Tham 2018 111) Two studies provided retrospective cohort data in adjusted form only, not amenable to meta-analysis.(Fink 2016 121) (Tijssen 2015 1)
All results are presented as Absolute Risk Difference (aRD) and our assessment of statistical significance is based on the absolute risk reduction 95% confidence interval (CI 95%). Relative Risk calculations were considered less informative and sometimes produced divergent results, likely as a consequence of zero-numerator cells. (Friedrich 2007 5) For reasons of clarity and unambiguity we decided only to report these for the single studies (and not for the meta-analysis). However, values can be found in the GRADE Summary of Findings tables that supplement the systematic review.
Survival with good neurologic outcome
For this critical outcome, we identified low certainty evidence downgraded for indirectness (resuscitations conducted prior to 2000, when standard resuscitation was different than current practice) and imprecision from 1 study. (Gausche 2000 783) This study assigned 591 children with out-of-hospital cardiac arrest (OHCA) to TI or BMV and found no statistical benefit or harm associated with use of TI (RR 0.69, 95% confidence interval (CI): 0.32-1.52; absolute risk difference aRD: -1.5% or 15 fewer children surviving with good neurologic outcome per 1,000 randomised to TI; CI 95%: 48 fewer to 17 more).
Additional, very low certainty evidence comes from three propensity-adjusted cohort studies, downgraded because none could differentiate children with failed attempts at TI from those in whom TI was not attempted (risk of bias). (Andersen 2016 1786) (Hansen 2017 51) (Ohashi-Fukuda 2017 66) These studies included 3,855 children with in-hospital cardiac arrest (IHCA) or OHCA and found apparent harm associated with the TI intervention (aRD: -4.9% or 49 fewer survivors per 1,000 resuscitations; CI 95%: 77 fewer to 21 fewer). In addition, 3 cohort studies involving 1,188 children had very heterogeneous results (I2: 95%).(del Castillo 2015 340) (Sirbaugh 1999 174) (Tham 2018 111)
Survival to hospital discharge
For this critical outcome, we identified low certainty evidence from one clinical trial of 591 children.(Gausche 2000 783) No apparent association was found between TI and survival (RR 1.04, CI 95%: 0.60-1.79; aRD: 0.3% or 3 more patients surviving to discharge per 1,000 assigned to TI; CI 95%: 41 fewer to 47 more).
Very low certainty evidence is provided by three propensity-adjusted cohort studies including 4,155 children. (Andersen 2016 1786) (Hansen 2017 51) (Ohashi-Fukuda 2017 66) Taken together, these studies suggested harm associated with TI (aRD: -5.3% or 53 fewer survivors per 1,000 patients; CI 95%: 86 fewer to 20 fewer). Combined data from 7 observational studies involving 4,539 patients were highly discordant (I2: 81%).(Abe 2012 612) (Aijian 1989 489) (Deasy 2010 1095) (Guay 2004 373) (Pitetti 2002 283) (Sirbaugh 1999 174) (Tham 2018 111)
Finally, very low certainty evidence is available from 2 observational studies, not amenable to meta-analysis, involving 3,992 children, each of which found no statistical benefit or harm from TI (Fink 2016: adjusted odds ratio (aOR) 0.64, CI 95%: 0.37 – 1.13; Tijssen 2015: aOR 0.69 (CI 95%: 0.43 – 1.10).(Fink 2016 121) (Tijssen 2015 1)
Survival to hospital admission
For this important outcome, no clinical trial data are available. Very low certainty evidence is provided by 1 propensity-matched cohort study of 1,508 patients, which found no statistical benefit or harm associated with TI (RR 0.99, CI 95%: 0.83 – 1.17; aRD: -0.3% or 3 fewer survivors per 1000 when TI is employed, CI 95%: 47 fewer to 41 more).(Hansen 2017 51) Three retrospective cohort studies involving 1,040 children produced discordant results (I2: 82%).(Aijian 1989 489) (Pitetti 2002 283) (Tham 2018 111)
Return of Spontaneous Circulation (ROSC)
No clinical trial data are available for this important outcome. Very low certainty evidence is obtained from 3 propensity-matched cohort studies involving 4,155 children, which found no statistical benefit or harm from the use of TI (aRD: 1.2% or 12 more patients with ROSC achieved per 1,000 resuscitations with TI; CI 95%: 15 fewer to 39 more). (Andersen 2016 1786) (Hansen 2017 51) (Ohashi-Fukuda 2017 66) The results of two other cohort studies involving 1,064 children had highly dissimilar results (I2: 97%).(Sirbaugh 1999 174) (Tham 2018 111)
Summary of findings
These results suggest that resuscitation with tracheal intubation is not superior to BMV-based resuscitation for cardiac arrest in children for the critically important outcomes of survival to hospital discharge and survival to hospital discharge with good neurologic outcomes (with low to very low certainty). Some very low certainty evidence suggests the use of TI may be associated with harm.
Studies comparing supraglottic airway placement (SGA) intubation with bag mask ventilation alone (BMV):
Only 4 studies were identified comparing SGA with BMV. All were observational in design and were conducted in the OHCA setting. Two studies provided propensity-adjusted cohort data.(Hansen 2017 51) (Ohashi-Fukuda 2017 66) Two studies provided simple observational data.(Abe 2012 612) (Tham 2018 111) Again, all results are presented as Absolute Risk Difference (aRD) and our assessment of statistical significance is based on the absolute risk reduction 95% confidence interval (CI 95%) (see above).
Survival with good neurologic outcome
For this critical outcome, very low certainty evidence obtained from two propensity-adjusted cohort studies involving 1,657 patients, showed no statistical benefit or harm associated with SGA ventilation (aRD: -2.9% or 29 fewer survivors with good neurologic outcome per 1,000 patients managed with SGA; CI 95%: 75 fewer to 17 more).(Hansen 2017 51) (Ohashi-Fukuda 2017 66) Additional very low certainty evidence comes from a non-adjusted observational study of 900 children. (Tham 2018 111) Like the previous studies, no significant benefit or harm was apparent (RR 0.75, CI 95%: 0.23 – 2.42; aRD: -0.9% or 9 fewer survivors per 1,000 resuscitations; CI 95%: 43 fewer to 24 more).
Survival to hospital discharge
For this critical outcome, the results of 2 propensity-adjusted cohort studies of 1,657 patients were highly discrepant (I2: 81%). (Hansen 2017 51) (Ohashi-Fukuda 2017 66)
Additional very low certainty evidence from two observational studies of 3,904 children found no significant treatment association (aRD: -3.5% or 35 fewer survivors per 1,000 treated with SGA; CI 95%: 88 fewer to 18 more).(Abe 2012 612) (Tham 2018 111)
Survival to hospital admission
For this important outcome, based on very low certainty evidence from a single propensity-matched cohort study of 996 children, no significant benefit or harm was associated with SGA (RR 1.25, CI 95%: 0.99-1.57; aRD: 6.4% or 64 more survivors to admission per 1,000 SGA resuscitations; CI 95%: 6 fewer to 133 more).(Hansen 2017 51) Additional very low certainty evidence is provided by a cohort study of 900 children, also showing no benefit or harm (RR 0.85, CI 95%: 0.44 – 1.87; aRD: -1.5% or 15 fewer patients surviving to admission/1000 treated with SGA; CI 95%: 70 fewer to 41 more).(Tham 2018 111)
Return of spontaneous circulation
For this important outcome, we identified very low certainty evidence from 2 propensity-matched cohort studies involving 1,657 children, which produced divergent results (I2: 84%).(Hansen 2017 51)(Ohashi-Fukuda 2017 66) Additional evidence of very low certainty can be obtained from a cohort study involving 900 children, in which no significant benefit or harm was found (RR 1.26, CI 95%: 0.82-1.92; aRD: 4% or 40 more patients achieving ROSC per 1,000 resuscitations with SGA; CI 95%: 41 fewer to 121 more).(Tham 2018 111)
Summary of findings
Recognizing the fact that there are conflicting and uncertain study results, the overall data are most consistent with no treatment effect associated with SGA ventilation when compared with BMV.
Studies comparing tracheal intubation (TI) with supraglottic airway placement (SGA):
The same 4 studies compared outcomes obtained with TI to those with SGA-based ventilation during resuscitation. All were conducted in the OHCA setting. Here too, all results are presented as Absolute Risk Difference (aRD) and our assessment of statistical significance is based on the absolute risk reduction 95% confidence interval (CI 95%) (see above).
Survival with good neurologic outcome
For this critical outcome, very low certainty evidence is available from 2 propensity-matched cohort studies enrolling 1,288 children.(Hansen 2017 51) (Ohashi-Fukuda 2017 66) When combined, these studies showed no statistical benefit or harm to either intervention (aRD: -2.2% or 22 fewer neurologically intact survivors per 1,000 patients managed with TI rather than SGA; CI 95%: 51 fewer to 6 more). Additional very low certainty evidence is provided by a cohort study of 127 patients, which also found no statistical advantage to either modality (RR 6.06, CI 95%: 1.32-27.7; aRD: 13.9% or 139 more survivors with TI; CI 95%: 36 fewer to 314 more).(Tham 2018 111)
Survival to hospital discharge
Similar to the above, for this critical outcome, very low certainty evidence from 2 propensity-matched cohort studies of 1,288 patients found no statistical benefit or harm associated with TI or SGA (aRD: -3.1% or 31 fewer children surviving per 1,000 resuscitations using TI compared with SGA; CI 95%: 73 fewer to 11 more). (Hansen 2017 51) (Ohashi-Fukuda 2017 66) Similar results were obtained from 2 cohort studies involving 582 children (aRD: 3.4% or 34 more survivors per 1,000 children managed with TI; CI 95%: 6 fewer to 75 more).(Abe 2012 612) (Tham 2018 111)
Survival to hospital admission
For this important outcome, very low certainty evidence from a propensity-matched cohort study found no statistical benefit or harm to either intervention (RR 0.79, CI 95%: 0.63-1.00; aRD: -6.7% or 67 fewer patients surviving to hospital admission per 1,000 managed with TI, as opposed to SGA; CI 95%: 136 fewer to 4 more).(Hansen 2017 51) In contrast, very low certainty evidence from a single cohort study found improved outcomes associated with TI (RR 4.33, CI 95%: 2.28-8.2; aRD: 47.2% or 472 more survivors per 1,000 resuscitations when TI is used rather than SGA; CI 95%: 198 more to 665 more).(Tham 2018 111)
Return of spontaneous circulation
For this important outcome, very low certainty evidence from 2 propensity-matched cohort studies involving 1,288 children, showed no statistical association between airway management choice and outcome (aRD: -2.6% or 26 fewer patients achieving ROSC per 1,000 managed with TI rather than SGA; CI 95%: 129 fewer to 78 more).(Hansen 2017 51) (Ohashi-Fukuda 2017 66) In contrast, very low certainty evidence from a single cohort study of 127 children found more patients achieving ROSC when TI was employed (RR 3.42, CI 95%: 2.16-5.44; aRD: 51.1% or 511 more patients achieving ROSC per 1,000 resuscitations with TI; CI 95%: 291 more to 732 more).(Tham 2018 111)
Summary of findings
There are no significant differences in outcomes shown between the use of TI or SGA in pediatric resuscitation based on limited and contradictory evidence of very low certainty.
Subgroup Analyses
In-hospital cardiac arrest (IHCA) vs. out-of-hospital cardiac arrest (OHCA)
Separate analyses of studies of IHCA and OHCA produced similar results. However, the body of evidence for IHCA is particularly small (consisting of 1 propensity-matched cohort study and 2 other cohort studies) and provides very low certainty evidence. (Andersen 2016 1786) (del Castillo 2015 340) (Guay 2004 373)
Traumatic vs. medical causes of cardiac arrest.
Six studies either excluded patients with traumatic causes of cardiac arrest or included fewer than 10 percent of such patients.(Hansen 2017 51) (Aijian 1989 489) (Deasy 2010 1095) (Guay 2004 373) (Fink 2016 121) (Tijssen 2015 1) The results of these studies are congruent with the overall results reported above. No study could be found focusing on patients with traumatic cardiac arrest.
Treatment recommendations
We suggest the use of BMV rather than TI or SGA in the management of children during cardiac arrest in the out-of-hospital setting (weak recommendation, very low certainty evidence).
We can make no recommendation about the use of TI or SGA in the management of children with cardiac arrest in the in-hospital setting owing to limited evidence
Justification and Evidence to Decision Framework Highlights
Advanced airway (AAW) interventions, particularly TI, have been long-established components of the advanced life support bundle of care in adults and children. As a result of inherent limitations in their design and data sources, the available studies, though individually well conducted, can provide only very low certainty evidence about whether attempting AAW placement prior to ROSC improves resuscitation outcomes. The best available data show no benefit from AAW interventions, and some suggested association with harm, for the critical outcomes of survival with good neurologic outcome and survival to hospital discharge. Placement of an AAW appears to be neutral for the short-term resuscitation outcomes of survival to hospital admission and ROSC. While these short-term outcomes do not ultimately benefit the patient, they may benefit the family, albeit at a monetary cost.
Effective BMV, TI, and SGA are all difficult skills that require good initial training, retraining, and quality control to be done consistently, safely, and effectively. Pediatric AAW programs require a moderate investment in equipment and a significant investment in training, skills maintenance, and quality control programs to be successful.
The benefit or harm associated with AAW-based resuscitation may differ across settings. Importantly, the available data do not inform the questions of whether better outcomes might be achieved by AAW-based strategies in long distance transportation or in prolonged resuscitation situations, with highly experienced airway operators. The analyzed data are only relevant to AAW interventions during CPR, and do not pertain to airway management in other critical situations.
Knowledge gaps
The only clinical trial undertaken in this area (Gausche 2000 783) was before the major changes in standard resuscitation practice that emphasise the need to minimise interruptions during chest compressions. There have been no clinical trials addressing airway management during cardiac arrest in the in-hospital setting.
Prehospital, ED-based, and in-hospital studies of similar design, ideally comparing TI, SGA, and BMV with planned subgroup analyses based on age and etiology of arrest (trauma vs non-trauma) are ethical, necessary, and critically important.
Attachments
Acknowledgement
The PLS Task Force would like to acknowledge the work undertaken by CEE and by AHA staff, in particular the assistance given by Dr. Matt Buchanan.
References
Abe T, Nagata T, Manabu H, Hagihara A. Life support techniques related to survival after out-of-hospital cardiac arrest in infants. Resuscitation. 2012;83:612 - 8.
Aijian P, Tsai A, Knopp R, Kallsen GW. Endotracheal intubation of pediatric patients by paramedics. Ann Emerg Med. 1989;18(5):489-94.
Andersen LW, Raymond TT, Berg RA, Nadkarni VM, Grossestreuer AV, Kurth T, et al. Association Between Tracheal Intubation During Pediatric In-Hospital Cardiac Arrest and Survival. JAMA. 2016;316(17):1786-97.
Deasy C, Bernard SA, Cameron P, Jaison A, Smith K, Harriss LR, et al. Epidemiology of paediatric out-of-hospital cardiac arrest in Melbourne, Australia. Resuscitation. 2010;81:1095 - 100.
del Castillo J, López-Herce J, Matamoros M, Cañadas S, Rodriguesz-Calvo A, Cecchetti C, et al. Long-term evolution after in-hospital cardiac arrest in children: Prospective multicenter multinational study. Resuscitation. 2015;96:126 - 34.
Fink EL, Prince DK, Kaltman JR, Atkins DL, Austin M, Warden C, et al. Unchanged pediatric out-of-hospital cardiac arrest incidence and survival rates with regional variation in North America. Resuscitation. 2016;107:121-8.
Friedrich JO, Adhikari NK, Beyene J. Inclusion of zero total event trials in meta-analyses maintains analytic consistency and incorporates all available data. BMC medical research methodology 2007;7:5.
Gausche M, Lewis RJ, Stratton SJ, Haynes BE, Gunter CS, Goodrich SM, et al. Effect of out-of-hospital pediatric endotracheal intubation on survival and neurological outcome: A controlled clinical trial. JAMA. 2000;283(6):783-90.
Guay J, Lortie L. An evaluation of pediatric in-hospital advanced life support interventions using the pediatric Utstein guidelines: a review of 203 cardiorespiratory arrests. Can J Anaesth. 2004;51(4):373-8.
Hansen ML, Lin A, Eriksson C, Daya M, McNally B, Fu R, et al. A comparison of pediatric airway management techniques during out-of-hospital cardiac arrest using the CARES database. Resuscitation. 2017;120:51-6.
Ohashi-Fukuda N, Fukuda T, Doi K, Morimura N. Effect of prehospital advanced airway management for pediatric out-of-hospital cardiac arrest. Resuscitation. 2017;114:66-72.
Pitetti R, Glustein JZ, Bhende MS. Prehospital care and outcome of pediatric out-of-hospital cardiac arrest. Prehosp Emerg Care. 2002;6(3):283-90.
Sirbaugh PE, Pepe PE, Shook JE, Kimball KT, Goldman MJ, Ward MA, et al. A prospective, population-based study of the demographics, epidemiology, management, and outcome of out-of-hospital pediatric cardiopulmonary arrest. Ann Emerg Med. 1999;33:174-84.
Tham LP, Wah W, Phillips R, Shahidah N, Ng YY, Shin SD, et al. Epidemiology and outcome of paediatric out-of-hospital cardiac arrests: A paediatric sub-study of the Pan-Asian resuscitation outcomes study (PAROS). Resuscitation. 2018;125:111-7.
Tijssen JA, Prince DK, Morrison LJ, Atkins DL, Austin MA, Berg R, et al. Time on the scene and interventions are associated with improved survival in pediatric out-of-hospital cardiac arrest. Resuscitation. 2015;94:1-7.