Skip to main content

Foundational Trials

CAAM

Image: Shutterstock / ChaNaWiT

Publication

  • Title: Effect of Bag-Mask Ventilation vs Endotracheal Intubation During Cardiopulmonary Resuscitation on Neurologic Outcome After Out-of-Hospital Cardiorespiratory Arrest: A Randomized Clinical Trial
  • Acronym: CAAM
  • Year: 2018
  • Journal published in: JAMA
  • Citation: Jabre P, Penaloza A, Pinero D, et al. Effect of Bag-Mask Ventilation vs Endotracheal Intubation During Cardiopulmonary Resuscitation on Neurologic Outcome After Out-of-Hospital Cardiorespiratory Arrest: A Randomized Clinical Trial. JAMA. 2018;319(8):779-787.

Context & Rationale

  • Background
    • Airway and ventilation strategies during out-of-hospital cardiac arrest (OHCA) remain a core component of resuscitation “bundles”, yet high-quality comparative evidence has historically been limited.
    • Endotracheal intubation (ETI) provides a definitive airway but can be technically difficult during CPR, may interrupt chest compressions, and carries risks of unrecognised oesophageal intubation, mainstem intubation, and aspiration.
    • Bag-mask ventilation (BMV) is immediately deployable and may avoid intubation-related pauses, but can be associated with gastric insufflation, regurgitation, aspiration, and inadequate ventilation during prolonged resuscitation.
    • Before CAAM, much of the comparative evidence was observational and vulnerable to confounding by indication (e.g., sicker patients preferentially receiving advanced airways), producing heterogeneous and contested inferences.
    • Practice therefore varied markedly by EMS model (physician-led vs paramedic-led systems), training intensity, and local performance metrics.
  • Research Question/Hypothesis
    • In adults with non-traumatic OHCA with an initial non-shockable rhythm, does an initial strategy of BMV achieve non-inferior 28-day favourable neurological outcome compared with an initial strategy of ETI?
    • Non-inferiority margin: 1.0% absolute difference for favourable neurological outcome.
  • Why This Matters
    • Airway management is frequent, time-critical, and operator-dependent; small effects can translate into large population-level differences.
    • Systems need evidence to justify investment in ETI training/maintenance versus alternative airway strategies.
    • Clarifying safety trade-offs (e.g., regurgitation vs intubation misplacement) is essential for protocol design and quality assurance.

Design & Methods

  • Research Question: Among adults with non-traumatic OHCA and an initial non-shockable rhythm, is an initial BMV strategy non-inferior to an initial ETI strategy for favourable neurological outcome at 28 days?
  • Study Type: Multicentre, randomised, open-label, parallel-group, non-inferiority trial; stratified by centre with permuted blocks; prehospital physician-led advanced life support EMS (15 centres in France and 5 centres in Belgium).
  • Population:
    • Setting: prehospital OHCA treated by basic life support teams plus physician-staffed advanced life support units.
    • Inclusion: age ≥18 years; non-traumatic OHCA; initial rhythm non-shockable; no spontaneous breathing and requiring advanced airway management.
    • Key exclusions: foreign body/airway obstruction; trauma; major external bleeding; do-not-resuscitate status.
  • Intervention:
    • Initial airway strategy: bag-mask ventilation delivered by the emergency physician.
    • Rescue: endotracheal intubation permitted if BMV failed or in the setting of regurgitation/aspiration concerns (reported as rescue intubation in the trial flow).
  • Comparison:
    • Initial airway strategy: endotracheal intubation by the emergency physician during CPR.
    • Rescue: bag-mask ventilation used before and between intubation attempts and in the event of intubation failure.
  • Blinding: Clinicians were not blinded (procedural impossibility); 28-day neurological outcome was assessed by an investigator blinded to group allocation.
  • Statistics: Assuming 3.0% favourable neurological outcome in both groups, a total of 2000 patients were required to demonstrate non-inferiority with 80% power at a 1-sided 2.5% significance level using a 1.0% absolute non-inferiority margin; primary analysis was intention-to-treat with supportive per-protocol and modified intention-to-treat analyses.
  • Follow-Up Period: 28–35 days (primary endpoint assessed at day 28 within a prespecified window).

Key Results

This trial was not stopped early. Enrolment reached target sample size (2043 randomised; 2040 included in the intention-to-treat analysis).

Outcome Bag-mask ventilation Endotracheal intubation Effect p value / 95% CI Notes
Favourable neurological outcome at day 28 (CPC 1–2; intention-to-treat) 44/1018 (4.3%) 43/1022 (4.2%) Risk difference 0.11% 1-sided 97.5% CI −1.64% to ∞; P=0.11 (non-inferiority) Non-inferiority margin was −1.0%; non-inferiority was not demonstrated.
Survival at day 28 (intention-to-treat) 55/1018 (5.4%) 54/1022 (5.3%) Risk difference 0.1% 95% CI −1.8% to 2.1%; P=0.90 Absolute survival was low in both groups (population restricted to non-shockable rhythms).
Survival to hospital admission (intention-to-treat) 337/1018 (33.1%) 376/1022 (36.8%) Risk difference −3.7% 95% CI −7.7% to 0.3%; P=0.07 Borderline difference favouring ETI, without downstream survival separation at 28 days.
Return of spontaneous circulation (intention-to-treat) 289/1018 (28.4%) 339/1022 (33.2%) Risk difference −4.7% 95% CI −8.8% to −0.5%; P=0.03 Intermediate endpoint difference favouring ETI.
Airway management difficulty (safety population) 186/1028 (18.1%) 138/1030 (13.4%) Risk difference 4.7% 95% CI 1.5% to 7.9%; P=0.004 Composite procedural difficulty; open-label ascertainment may influence reporting.
Airway management failure (safety population) 69/1028 (6.7%) 22/1030 (2.1%) Risk difference 4.6% 95% CI 2.8% to 6.4%; P<0.001 In the BMV group, rescue intubation was performed in 146 patients (55 for ventilation failure and 100 for gastric regurgitation; not mutually exclusive).
Regurgitation (safety population) 156/1028 (15.2%) 77/1030 (7.5%) Risk difference 7.7% 95% CI 4.9% to 10.4%; P<0.001 Safety signal favouring ETI; clinical sequelae (e.g., aspiration pneumonitis) were not reported as patient-centred endpoints.
Recognised oesophageal intubation (safety population) Not applicable 105/1030 (10.2%) Not reported Not reported ETI-specific complication; recognised and corrected. Additional ETI complications: mainstem intubation 21/1030 (2.0%), dental injury 7/1030 (0.7%), accidental extubation 5/1030 (0.5%).
  • The primary endpoint was rare (≈4.2% favourable neurological outcome), and the trial did not meet its prespecified non-inferiority criterion for BMV.
  • Despite lower ROSC in the BMV group (28.4% vs 33.2%), 28-day survival and neurological outcomes were very similar between strategies.
  • Procedural harms differed by strategy: BMV was associated with more regurgitation (15.2% vs 7.5%) and airway failure (6.7% vs 2.1%), whereas ETI carried ETI-specific misplacement events (recognised oesophageal intubation 10.2%).

Internal Validity

  • Randomisation and allocation concealment: Randomisation was stratified by centre with permuted blocks; allocation used sequentially numbered, sealed opaque envelopes opened in the field after eligibility confirmation, supporting concealment at the point of assignment.
  • Post-randomisation exclusions: 3/2043 randomised patients were excluded from the intention-to-treat analysis (2 do-not-resuscitate; 1 not in cardiac arrest), a small proportion with limited impact on effect estimates.
  • Performance and detection bias: Treatment delivery could not be blinded; however, the primary neurological outcome (CPC at day 28) was assessed by a blinded investigator, reducing detection bias for the primary endpoint.
  • Protocol adherence and separation: Per-protocol population sizes differed (BMV 995 vs ETI 943), reflecting differential protocol deviations and crossovers; rescue intubation occurred in 146 BMV-assigned patients, and 14 ETI-assigned patients received BMV as the initial strategy.
  • Baseline comparability: Groups were well balanced (mean age 67.5 vs 68.7 years; male 67.1% vs 67.1%; bystander CPR 70.5% vs 72.7%; asystole 61.4% vs 60.9%; time from collapse to advanced life support 15.0 [IQR 10–23] vs 15.0 [IQR 10–23] minutes).
  • Timing and delivery context: Intervention was delivered in a physician-staffed prehospital system; the trial did not report quantitative measures of chest compression interruption attributable to airway manoeuvres, limiting mechanistic inference about “hands-off time”.
  • Outcome assessment: Primary endpoint (CPC) is clinically meaningful but relatively coarse and susceptible to practice variation in post-resuscitation care and rehabilitation access; nonetheless, it is standard in OHCA trials.
  • Statistical rigour: The trial achieved planned sample size for the non-inferiority hypothesis; the very low event rate for the primary outcome produced wide confidence bounds, and the lower bound crossed the prespecified non-inferiority margin.

Conclusion on Internal Validity: Overall, internal validity appears moderate: randomisation and blinded primary outcome assessment were strengths, but open-label delivery, substantial rescue/crossover in the BMV strategy, and a rare primary endpoint limit precision and interpretability of “strategy separation”.

External Validity

  • Population representativeness: The trial enrolled a restricted OHCA subgroup (non-traumatic arrests with initial non-shockable rhythm), resulting in low baseline survival and potentially limiting applicability to shockable rhythms and witnessed VF/VT populations.
  • System context: Results derive from physician-staffed prehospital advanced life support teams; operator skill, decision thresholds for rescue airway conversion, and suction/airway adjunct availability may differ substantially in paramedic-only or resource-limited systems.
  • Intervention generalisability: The trial compares initial airway strategies rather than mutually exclusive, sustained separation (given rescue intubation in BMV), which mirrors real-world “step-up” practice but complicates translation to rigid protocolised systems.

Conclusion on External Validity: Generalisability is moderate: findings are most applicable to physician-led EMS systems treating predominantly non-shockable OHCA, and less directly applicable to systems with different airway competencies, device availability, or patient case-mix (e.g., shockable rhythms).

Strengths & Limitations

  • Strengths: Large pragmatic prehospital randomised trial; centre-stratified randomisation with allocation concealment; clinically meaningful primary endpoint (28-day neurological outcome); blinded assessment of the primary outcome; prespecified non-inferiority framework with supportive sensitivity analyses.
  • Limitations: Open-label procedural trial; restricted to non-shockable rhythms (low event rate and limited applicability); frequent rescue airway conversion (146 rescue intubations in BMV group) reducing separation; airway-related mechanisms (e.g., compression interruption time, ventilation quality) not quantitatively reported; non-inferiority interpretation sensitive to chosen margin and low event frequency.

Interpretation & Why It Matters

  • Clinical meaning of “not non-inferior”
    The point estimates for 28-day favourable neurological outcome were essentially identical (4.3% vs 4.2%), but the prespecified non-inferiority criterion was not met due to confidence interval width; clinically, this supports equipoise rather than superiority of either initial strategy in this population.
  • Strategy selection is safety-trade-off driven
    BMV carried higher regurgitation and airway failure, prompting frequent rescue ETI; ETI carried non-trivial recognised oesophageal intubation (10.2%), underscoring that airway choice is inseparable from operator skill, confirmation practices, and readiness to “step up/down”.
  • Systems implication
    In physician-led EMS, a BMV-first approach requires robust suction, vigilance for regurgitation, and rapid access to definitive airway rescue, whereas an ETI-first approach requires stringent confirmation and mitigation of misplacement risk; neither strategy eliminates the need for high-reliability airway governance.

Controversies & Subsequent Evidence

  • Non-inferiority design sensitivity: With favourable neurological outcome ~4%, a 1% absolute non-inferiority margin is proportionally large; combined with rare events, the design is highly sensitive to small shifts in event counts and to protocol deviations.
  • Interpretation of crossover/rescue as “real-world strategy” vs “dilution”: Rescue intubation in 146 BMV-assigned patients indicates pragmatic realism (step-up practice), but reduces separation and complicates attribution of downstream effects to the initial airway strategy.
  • Generalisability debate: Physician-staffed ETI performance and the threshold for declaring BMV failure or regurgitation may not translate to paramedic systems, where supraglottic devices are often the default intermediate airway.
  • Relationship to other pragmatic airway RCTs: Two large 2018 prehospital randomised trials compared supraglottic airways with ETI and broadly found no clear functional superiority of ETI, while highlighting differences in airway success and complications: AIRWAYS-22 and PART3.
  • Guideline incorporation: Contemporary resuscitation guidelines emphasise choosing an airway strategy that minimises interruptions and matches provider competence, recognising BMV and supraglottic airways as appropriate early strategies, with ETI reserved for highly trained operators and/or specific clinical contexts (e.g., prolonged resuscitation, aspiration risk).45

Summary

  • CAAM randomised 2043 adults with non-traumatic OHCA and initial non-shockable rhythm to an initial BMV vs ETI strategy in physician-led EMS systems.
  • Favourable neurological outcome at 28 days was 4.3% (BMV) vs 4.2% (ETI); the trial did not meet its prespecified non-inferiority criterion for BMV.
  • ROSC was lower with BMV (28.4% vs 33.2%), but 28-day survival was similar (5.4% vs 5.3%).
  • Safety signals differed: BMV had more regurgitation (15.2% vs 7.5%) and airway failure (6.7% vs 2.1%); ETI had ETI-specific misplacement events (recognised oesophageal intubation 10.2%).
  • Interpretation is best framed as a pragmatic comparison of initial airway strategies with substantial rescue conversion, rather than a pure comparison of sustained exclusive airway modalities.

Further Reading

Other Trials

Systematic Review & Meta Analysis

Observational Studies

Guidelines

Notes

  • Where a DOI could not be confirmed from the provided documents, links are provided as PubMed search URLs to support verification and retrieval.

Overall Takeaway

CAAM is a landmark prehospital airway trial because it moved the field from confounded observational comparisons to a pragmatic, randomised evaluation of initial airway strategies in OHCA. In a high-skill, physician-led EMS context and a low-survival (non-shockable) population, BMV did not meet a strict non-inferiority criterion versus ETI for 28-day favourable neurological outcome, yet absolute patient-centred outcomes were nearly identical—shifting emphasis towards system-appropriate airway strategies that prioritise minimising interruptions and managing predictable procedure-specific harms.

Overall Summary

  • Randomised, pragmatic comparison of initial BMV vs ETI strategies in non-shockable OHCA treated by physician-led EMS.
  • Favourable neurological outcome at 28 days was ~4% in both groups; non-inferiority for BMV was not demonstrated.
  • Safety trade-offs differed: BMV had more regurgitation/failure; ETI had misplacement events (recognised oesophageal intubation 10.2%).

Bibliography

  • 1.Lewis RJ, Gausche-Hill M. Airway management during out-of-hospital cardiac arrest. JAMA. 2018;319(8):771-772. Link
  • 2.Benger JR, Kirby K, Black S, et al. Effect of a Strategy of a Supraglottic Airway Device vs Tracheal Intubation During Out-of-Hospital Cardiac Arrest on Functional Outcome: The AIRWAYS-2 Randomized Clinical Trial. JAMA. 2018;320(8):779-791. Link
  • 3.Wang HE, Schmicker RH, Daya MR, et al. Effect of a Strategy of Initial Laryngeal Tube Insertion vs Endotracheal Intubation on 72-Hour Survival in Out-of-Hospital Cardiac Arrest: The PART Randomized Clinical Trial. JAMA. 2018;320(8):769-778. Link
  • 4.Panchal AR, Bartos JA, Cabañas JG, et al. Part 3: Adult Basic and Advanced Life Support: 2020 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Circulation. 2020;142(16_suppl_2):S366-S468. Link
  • 5.Soar J, Böttiger BW, Carli P, et al. European Resuscitation Council Guidelines 2021: Adult advanced life support. Resuscitation. 2021;161:115-151. Link