Skip to main content

Foundational Trials

BEAM

Image: Shutterstock / ChaNaWiT

Publication

  • Title: Effect of Use of a Bougie vs Endotracheal Tube and Stylet on First-Attempt Intubation Success Among Patients With Difficult Airways Undergoing Emergency Intubation: A Randomized Clinical Trial
  • Acronym: BEAM (Bougie Use in Emergency Airway Management)
  • Year: 2018
  • Journal published in: JAMA
  • Citation: Driver BE, Prekker ME, Klein LR, Reardon RF, Miner JR, Fagerstrom ET. Effect of Use of a Bougie vs Endotracheal Tube and Stylet on First-Attempt Intubation Success Among Patients With Difficult Airways Undergoing Emergency Intubation: A Randomized Clinical Trial. JAMA. 2018;319(21):2179-2189.

Context & Rationale

  • Background
    • First-attempt success during emergency tracheal intubation is clinically important because repeated attempts are associated with hypoxaemia, hypotension, aspiration, and cardiac arrest.
    • Tracheal tube introducers (“bougies”) are widely used as rescue adjuncts when glottic visualisation is limited, but evidence for routine bougie-first use in emergency department (ED) intubation was limited and largely observational or simulation-based before BEAM.
    • In many EDs, the default first-pass technique remained an endotracheal tube preloaded with a malleable stylet, despite uncertainty about which approach maximises first-pass success in patients with anticipated difficulty.
  • Research Question/Hypothesis
    • In adults undergoing ED orotracheal intubation, does a bougie-first technique increase first-attempt intubation success compared with an endotracheal tube with stylet?
    • Hypothesis: bougie-first would improve first-attempt success, particularly among patients with difficult airway characteristics.
  • Why This Matters
    • If a low-cost, widely available adjunct meaningfully improves first-attempt success without increasing complications, it has immediate implications for ED airway training, standard operating procedures, and equipment layout.
    • The study also tests whether a “default bougie-first” strategy can outperform the prevailing “tube-with-stylet” approach under real-world ED constraints.

Design & Methods

  • Research Question: Among adults undergoing ED orotracheal intubation, does initial use of a bougie (vs an endotracheal tube preloaded with a stylet) increase first-attempt intubation success, particularly in those with at least 1 difficult airway characteristic?
  • Study Type: Randomised, single-centre, pragmatic clinical trial in a high-volume academic ED (Hennepin County Medical Center, Minneapolis, Minnesota); 1:1 patient-level allocation with block randomisation, stratified by obesity and cervical spine immobilisation; clinician-unblinded; ClinicalTrials.gov NCT02902146.
  • Population:
    • Setting: Adult ED intubations (September 2016 to August 2017), with planned Macintosh-style video laryngoscopy (Storz C-MAC) on the first attempt.
    • Inclusion: Age ≥18 years; clinician decision to perform emergency orotracheal intubation in the ED; availability of study envelope and equipment; intended use of a Macintosh-style blade.
    • Key exclusions: Age <18 years; prisoner; pregnancy; known or suspected anatomic distortion of the upper airway (eg, tumour, oedema, abscess, epiglottitis); need for immediate “crash” intubation without time for randomisation; non-Macintosh blade (including hyperangulated blades) planned for the first attempt.
    • Primary analysis cohort (pre-specified): Patients with ≥1 “difficult airway characteristic” documented by the intubator after the procedure (eg, blood/vomit in airway, cervical immobilisation, obesity, restricted mouth opening/neck movement, short neck, large tongue).
  • Intervention:
    • Bougie-first strategy: after laryngoscope insertion, a tracheal tube introducer was advanced first, then the endotracheal tube was advanced over the bougie into the trachea; subsequent steps (tube size, medications, confirmation methods) followed local standard practice.
    • Training and standardisation: clinicians received brief instruction and a standardised “bougie-first” approach consistent with the study protocol; the Macintosh-style video laryngoscope remained the default laryngoscopy device.
  • Comparison:
    • Endotracheal tube with stylet: a standard endotracheal tube preloaded with a malleable stylet (standard ED technique) was used for the initial attempt; subsequent attempts and rescue devices (including bougie) were at clinician discretion.
    • Co-interventions: choice of induction agent, neuromuscular blockade, pre-oxygenation method, and confirmation techniques were not protocolised and reflected routine ED practice.
  • Blinding: Unblinded (operators and bedside team); outcomes were primarily procedural and objectively verifiable (eg, end-tidal CO₂, oxygen saturation), with video review used when documentation was discrepant.
  • Statistics:
    • Power: a total of 374 patients with difficult airway characteristics were required to detect a 9% absolute increase in first-attempt success (from 86% to 95%) with 80% power at a two-sided α=0.05; pre-planned interim analyses for futility after enrolment of 250 and 500 participants.
    • Analysis: intention-to-treat; primary analysis used a χ² test for the primary outcome with sensitivity analyses accounting for clustering by clinician and for baseline covariates.
  • Follow-Up Period: Peri-intubation: through completion of the intubation attempt(s) and immediate complications recorded during and immediately after the procedure (including oxygen saturation nadir and haemodynamic events), with chart review for specified complications.

Key Results

This trial was not stopped early. It completed enrolment with pre-planned interim analyses for futility.

Outcome Bougie-first Endotracheal tube + stylet Effect p value / 95% CI Notes
First-attempt intubation success (primary; ≥1 difficult airway characteristic) 191/198 (96%) 150/182 (82%) +14% 95% CI 8% to 20%; P<0.001 Pre-specified primary cohort based on difficult airway characteristics documented after intubation
First-attempt intubation success (all patients) 373/381 (98%) 328/376 (87%) +11% 95% CI 7% to 14%; P<0.001 Intention-to-treat
First-attempt intubation success (no difficult airway characteristic) 182/183 (99%) 178/194 (92%) +7% 95% CI 2% to 12%; P=0.004 Planned secondary outcome
First-attempt success without hypoxaemia (SpO₂ <90% at any point) 290/381 (76%) 251/376 (67%) +9% 95% CI 3% to 16%; P=0.002 Composite efficacy–safety endpoint
Duration of first intubation attempt, seconds (median [IQR]) 38 (24–59) 36 (25–53) +2 s 95% CI −2 to 5; P=0.18 No meaningful difference in speed
Composite of intubation complications 66/381 (17%) 64/376 (17%) 0% 95% CI −5% to 6%; P=0.83 Included hypoxaemia, severe hypoxaemia, aspiration/regurgitation, oesophageal intubation, hypotension, cardiac arrest, airway trauma, laryngospasm
Oesophageal intubation 0/381 (0%) 3/376 (1%) −1% 95% CI −2% to 0%; P=0.08 Rare event; CI includes no effect
Severe hypoxaemia (SpO₂ <80%) 33/381 (9%) 30/376 (8%) +1% 95% CI −3% to 5%; P=0.59 No statistically significant difference
  • Unplanned subgroup analyses in the difficult-airway cohort showed similar direction of effect for blood/vomit in airway (28/30 [93%] vs 13/20 [65%]; absolute difference 28%; 95% CI 8% to 48%; P=0.01), cervical immobilisation (46/48 [96%] vs 38/53 [72%]; difference 24%; 95% CI 9% to 39%; P=0.003), and obesity (132/138 [96%] vs 71/92 [77%]; difference 18%; 95% CI 9% to 28%; P<0.001).
  • First-device contamination occurred: bougie was the first device inserted in 79/376 (21%) of the stylet group, yet first-pass success remained higher in the bougie-first arm.
  • Among 56 patients without first-pass success, the bougie was the successful technique in 49 (88%) on subsequent attempts.

Internal Validity

  • Randomisation and allocation concealment: 1:1 assignment using sequentially numbered opaque envelopes opened immediately before intubation; block randomisation (block sizes 2–10) stratified by obesity and cervical spine immobilisation.
  • Dropout/exclusions after randomisation were minimal for the primary procedural endpoint; 757 patients were randomised and included in intention-to-treat analyses, with 380 classified as having ≥1 difficult airway characteristic (bougie group n=198; stylet group n=182).
  • Performance/detection bias: operators were not blinded to device assignment; however, the primary outcome (first-attempt tracheal intubation success) and key safety outcomes (oxygen saturation nadir, hypotension, cardiac arrest) are largely objective and were recorded contemporaneously, with video review for discrepancies.
  • Protocol adherence and separation: adherence to assigned technique was high (bougie group 373/381 [98%] vs stylet group 346/376 [92%]); first device inserted after laryngoscope was a bougie in 369/381 (97%) vs 79/376 (21%), indicating meaningful (but incomplete) group separation.
  • Crossover/contamination: 79/376 (21%) of the stylet arm used a bougie as the first device inserted, which would be expected to bias the estimated treatment effect towards the null.
  • Baseline characteristics: groups were similar at baseline (eg, mean age 46 vs 45 years; mean pre-intubation oxygen saturation 97% vs 96%; median Glasgow Coma Scale 7 in both groups), supporting comparability.
  • Heterogeneity: single-centre setting but a broad range of ED indications and operators; sensitivity analyses accounting for clinician clustering and baseline covariates were reported and were directionally consistent.
  • Timing and dose: the intervention is procedural and delivered at the first attempt; attempt duration was similar (median 38 [IQR 24–59] vs 36 [25–53] seconds), suggesting the effect on first-pass success was not explained by longer attempts.
  • Outcome assessment and definitions: “difficult airway characteristics” were defined a priori but documented by the operator after the procedure, creating potential for subjective misclassification; primary analysis was nevertheless pre-specified.
  • Statistical rigour: the primary cohort size met the planned requirement (380 vs 374); intention-to-treat analysis was primary; interim futility analyses were planned.

Conclusion on Internal Validity: Overall, internal validity appears moderate-to-strong: randomisation and allocation concealment were robust and outcomes were objective, but the unblinded design, post-randomisation classification of the primary cohort, and control-group contamination introduce residual risk of bias.

External Validity

  • Population representativeness: adult ED patients undergoing emergency orotracheal intubation with a Macintosh-style video laryngoscope; common ED indications (medical and trauma) were represented.
  • Key exclusions limit applicability to: paediatric patients; prisoners; pregnancy; known/suspected anatomic upper-airway distortion; and “crash” intubations where immediate action precluded randomisation.
  • Device and technique context: the study was embedded in a setting with high baseline use of Macintosh-style video laryngoscopy; results may differ where direct laryngoscopy predominates, where different blades are used (eg, hyperangulated), or where bougie familiarity is low.
  • Feasibility and scalability: bougies are inexpensive and broadly available; implementation depends chiefly on training and integration into local airway checklists and standard operating procedures.

Conclusion on External Validity: Generalisability is moderate to similar high-volume EDs using Macintosh-style video laryngoscopy; extrapolation to crash intubations, markedly distorted upper airways, and other settings (eg, ICU) should be cautious.

Strengths & Limitations

  • Strengths:
    • Randomised pragmatic design with patient-level allocation in a real-world ED airway environment.
    • Large ED airway trial sample (n=757) with a pre-specified primary cohort of difficult airways (n=380).
    • High protocol adherence and clear procedural endpoints.
    • Objective, clinically relevant outcomes (first-pass success; hypoxaemia; hypotension; cardiac arrest) with video review for discrepancies.
  • Limitations:
    • Single-centre academic ED; practice patterns (eg, high use of Macintosh-style video laryngoscopy) may not generalise.
    • Clinicians were unblinded, with potential for performance bias (eg, differential optimisation, patience, or rescue decisions).
    • Primary analysis cohort (“difficult airway characteristics”) was determined after randomisation and depended on operator documentation, creating risk of subjective misclassification.
    • Meaningful contamination: 21% of the stylet arm had a bougie inserted first, potentially diluting between-group differences.
    • Peri-intubation outcomes were emphasised; downstream patient-centred outcomes (eg, ventilator days, mortality) were not reported.

Interpretation & Why It Matters

  • Practical signal
    Bougie-first increased first-attempt success in the pre-specified difficult-airway cohort (191/198 [96%] vs 150/182 [82%]; absolute difference 14%; 95% CI 8% to 20%; P<0.001) without increasing the composite complication rate (17% vs 17%).
  • Mechanism and workflow
    The bougie-first approach plausibly mitigates restricted tube passage when the laryngeal inlet is only partially visualised; similar attempt durations (median 38 vs 36 seconds) suggest the strategy improved success rather than merely prolonging attempts.
  • Implementation
    Adoption is primarily a training and systems question: ensuring immediate bougie availability, standardised technique, and clear rescue pathways when first-attempt success fails.

Controversies & Subsequent Evidence

  • Correspondence emphasised that BEAM was single-centre and embedded in a specific airway culture (high Macintosh-style video laryngoscopy use); it also questioned whether operator-reported “difficult airway characteristics” (recorded after the procedure) could introduce subjective classification that complicates interpretation of the primary cohort analysis. 12
  • Contamination in the control group (bougie inserted first in 79/376 [21%]) raises a plausible bias towards the null; this makes the observed first-pass benefit notable, but it also underscores that pragmatic trials of procedural technique can be vulnerable to clinician preference and rescue behaviour.
  • Subsequent multicentre evidence in critically ill adults (the JAMA 2021 BOUGIE trial) did not replicate a first-pass success advantage for bougie-first (80.4% vs 83.0%; absolute difference −2.6%; 95% CI −7.3% to 2.2%; P=0.27) and reported numerically more severe hypoxaemia with bougie-first (11.0% vs 8.8%; absolute difference 2.2%; 95% CI −1.6% to 6.0). Differences in setting, operator mix, physiology, and airway urgency are plausible effect modifiers. 3
  • A 2024 systematic review and meta-analysis synthesised trial evidence on bougie use and first-attempt success across settings and provides a framework for interpreting between-setting heterogeneity (ED vs ICU vs operating theatre). 4
  • Recent airway guidance emphasises first-pass success as a systems objective (optimised pre-oxygenation, physiologic preparation, and structured rescue plans) and incorporates introducers/bougies as important adjuncts within difficult-airway strategies rather than as a stand-alone solution. 56

Summary

  • BEAM randomised 757 adults undergoing ED emergency orotracheal intubation to bougie-first vs endotracheal tube with stylet.
  • The primary outcome (first-attempt success in the pre-specified difficult-airway cohort) favoured bougie-first: 191/198 (96%) vs 150/182 (82%); absolute difference 14%; 95% CI 8% to 20%; P<0.001.
  • First-attempt success in all patients also improved: 373/381 (98%) vs 328/376 (87%); absolute difference 11%; 95% CI 7% to 14%; P<0.001.
  • Despite improved first-attempt success, peri-intubation complications were similar (composite 17% vs 17%); no statistically significant difference in severe hypoxaemia, airway trauma, or cardiac arrest was reported.
  • The trial supports bougie-first as a pragmatic ED strategy to improve first-pass success, but subsequent ICU trial data and setting-specific expertise temper assumptions of universal benefit.

Further Reading

Other Trials

Systematic Review & Meta Analysis

Observational Studies

Guidelines

Notes

  • Key caution for interpretation: the effect of bougie-first strategies appears context-sensitive (ED vs ICU vs operating theatre) and is likely modified by operator experience, laryngoscopy device choice, and the physiology of the intubated population.

Overall Takeaway

BEAM demonstrated a large, clinically meaningful increase in first-pass intubation success with a bougie-first strategy during ED emergency intubation, including in a pre-specified cohort with difficult airway characteristics, without an observed increase in peri-intubation complications. The trial is landmark because it operationalised an inexpensive adjunct as a default first-pass strategy and forced the field to reconcile pragmatic ED benefits with mixed signals from subsequent ICU trials, sharpening focus on setting-specific airway systems and training.

Overall Summary

  • Bougie-first improved ED first-pass success (primary cohort absolute difference 14%) without increasing measured peri-intubation harms.
  • Control-group contamination was substantial (21% received bougie first) yet the benefit persisted, supporting a real technique effect in this setting.
  • Subsequent ICU trial data did not replicate benefit, reinforcing that airway interventions are highly context dependent.

Bibliography