Skip to main content

Foundational Trials

STYLETO

Image: Shutterstock / ChaNaWiT

Publication

  • Title: Effect of the use of an endotracheal tube and stylet versus an endotracheal tube alone on first-attempt intubation success: a multicentre, randomised clinical trial in 999 patients
  • Acronym: STYLETO
  • Year: 2021
  • Journal published in: Intensive Care Medicine
  • Citation: Jaber S, Rollé A, Godet T, Terzi N, Riu B, Asfar P, et al. Effect of the use of an endotracheal tube and stylet versus an endotracheal tube alone on first-attempt intubation success: a multicentre, randomised clinical trial in 999 patients. Intensive Care Med. 2021;47:653-664.

Context & Rationale

  • Background
    • Tracheal intubation in the ICU is frequently performed under physiological instability, where multiple attempts are associated with hypoxaemia, haemodynamic collapse, and peri-intubation cardiac arrest.
    • Adjuncts to increase first-pass success (FPS) are attractive because they are low-cost and broadly available, but their net benefit in the ICU (efficacy, complications, and trauma) was uncertain.
    • Routine stylet use is common in theatre practice, but ICU evidence had largely been extrapolated from non-ICU settings and non-randomised data.
  • Research Question/Hypothesis
    • Does routine first-attempt use of a straight-to-cuff stylet (with direct laryngoscopy) increase FPS compared with an endotracheal tube alone in critically ill adults requiring intubation for invasive ventilation?
    • Is any gain in FPS achieved without an increase in intubation-related complications or airway trauma?
  • Why This Matters
    • Even modest improvements in FPS could be clinically important given the high baseline risk of physiological complications in ICU intubation.
    • If safe and effective, a stylet-first approach could be a scalable, low-resource optimisation that complements broader “intubation bundles”.
    • Clarifying benefit (and harms) informs guideline recommendations about routine adjunct use during emergency ICU intubation.

Design & Methods

  • Research Question: In critically ill adults requiring orotracheal intubation for invasive mechanical ventilation, does first-attempt intubation using an endotracheal tube plus stylet improve FPS compared with an endotracheal tube alone, and what is the effect on intubation-related complications?
  • Study Type: Multicentre, parallel-group, randomised clinical trial in 32 French ICUs; pragmatic delivery alongside an ICU intubation bundle; operator unblinded with efforts to blind research staff/outcome recording; investigator-initiated. 1
  • Population:
    • Setting: adult ICUs (France); emergency and non-emergency intubations for invasive ventilation.
    • Key inclusion: adults requiring orotracheal intubation for invasive mechanical ventilation.
    • Key exclusions: cardiac arrest at the time of intubation; prior trial inclusion; “protected” persons; refusal/withdrawal of consent; other trial-specified exclusions.
  • Intervention:
    • First attempt performed with a standard endotracheal tube preloaded with a straight-to-cuff stylet (distal bend approximately 25–35°) during direct laryngoscopy.
    • Subsequent attempts/rescue devices permitted according to local practice after first-attempt failure.
  • Comparison:
    • First attempt performed with an endotracheal tube alone (no stylet) during direct laryngoscopy.
    • Stylet and other adjuncts allowed for subsequent attempts/rescue after first-attempt failure.
  • Blinding: Operators were necessarily unblinded; allocations were concealed until assignment, and peri-intubation data collection was designed to be performed by research staff not performing the intubation (risk reduction for detection bias for some outcomes, but not complete blinding for all elements).
  • Statistics: Sample size 1040 planned to detect a 10% absolute increase in FPS (70% to 80%) with 95% power at the 5% two-sided significance level (allowing <10% missing); analysis primarily by intention-to-treat (modified ITT for those with no intubation data). 1
  • Follow-Up Period: Procedural outcomes immediately and within 1 hour; exploratory physiological outcomes up to 24 hours; clinical outcomes (ICU metrics) and mortality to 90 days.

Key Results

This trial was not stopped early. Recruitment reached the planned randomised sample (1040), with the primary analysis conducted in a modified ITT cohort of 999 patients with intubation data.

Outcome Tracheal tube + stylet Tracheal tube alone Effect p value / 95% CI Notes
Primary: first-attempt intubation success 392/501 (78.2%) 356/498 (71.5%) RR 1.10 95% CI 1.02 to 1.18; P=0.01 Absolute difference 6.7% (95% CI 1.4 to 12.1); NNT 14.8 (95% CI 8.3 to 71.7).
Main secondary: complications related to intubation within 1 hour 194/501 (38.7%) 200/498 (40.2%) RR 0.96 95% CI 0.83 to 1.12; P=0.64 Absolute difference −1.5% (95% CI −7.5 to 4.6).
Severe complications (within 1 hour) 128/501 (25.6%) 128/498 (25.7%) RR 0.99 95% CI 0.81 to 1.23; P=1.00 Includes severe hypoxaemia and severe haemodynamic collapse components.
Operator-assessed difficult intubation 86/501 (17.2%) 116/498 (23.3%) RR 0.74 95% CI 0.57 to 0.95; P=0.15 Exploratory; P adjusted (false discovery rate).
Oesophageal intubation 10/501 (2.0%) 12/498 (2.4%) RR 0.83 95% CI 0.36 to 1.90; P=0.97 Component of moderate complications.
Operator-reported aspiration 13/501 (2.6%) 5/498 (1.0%) RR 2.58 95% CI 0.93 to 7.19; P=0.33 Exploratory; wide CI; P adjusted (false discovery rate).
Traumatic injuries (serious adverse events) 20/501 (4.0%) 18/498 (3.6%) RR 1.10 95% CI 0.59 to 2.06; P=0.76 Includes mucosal bleeding; laryngeal, tracheal, mediastinal, and oesophageal injuries were rare.
28-day mortality 158/501 (31.5%) 150/498 (30.1%) RR 1.05 95% CI 0.87 to 1.26; P=0.99 Exploratory clinical outcome; P adjusted (false discovery rate).
90-day mortality 180/501 (35.9%) 188/498 (37.8%) RR 0.95 95% CI 0.81 to 1.12; P=1.00 Exploratory clinical outcome; P adjusted (false discovery rate).
  • Routine stylet use increased FPS by 6.7% (78.2% vs 71.5%), but did not reduce the composite of intubation-related complications within 1 hour (38.7% vs 40.2%).
  • Severe complications were common (~26%) and were not reduced by the intervention; traumatic injuries were uncommon (~4%) and similar between groups.
  • Exploratory endpoints (including ICU-free days, ventilator-free days, and 28- and 90-day mortality) did not differ between groups.

Internal Validity

  • Randomisation and allocation: Central randomisation with allocation concealment until assignment; multicentre implementation reduced centre-specific selection bias risk.
  • Dropout/exclusions: 1040 randomised; modified ITT analysed 999 (stylet 501; control 498); exclusions after randomisation were balanced (21 vs 20), including improved clinical status (4 vs 5), consent withdrawal (8 vs 9), prior inclusion (6 vs 1), “protected persons” (3 vs 5), and death before intubation (0 vs 5).
  • Performance/detection bias: Operators unblinded; peri-intubation data collection was structured to be performed by non-operator research staff, but subjective components (e.g., operator-reported aspiration; perceived difficulty) remain at risk of bias.
  • Protocol adherence: High uptake of the accompanying ICU intubation bundle elements (e.g., rapid sequence induction 97.6% vs 97.6%; two-operator approach 94.9% vs 94.7%; non-invasive ventilation preoxygenation 59.3% vs 56.4%).
  • Baseline characteristics: Groups were clinically similar (age 63 ± 15 vs 62 ± 15 years; SAPS II 46 ± 18 vs 47 ± 20; SOFA 6.0 ± 3.6 vs 5.9 ± 3.7; expert operator 23.0% vs 23.5%).
  • Heterogeneity: 32 ICUs with varied operator backgrounds; pre-specified subgroup analyses did not show a clear interaction with operator expertise, call status, hypoxaemia, or predicted difficulty (not powered for interaction testing).
  • Timing: Intervention applied at the first laryngoscopy attempt; pragmatic timing consistent with emergency ICU practice.
  • Dose (intervention fidelity): Stylet use was protocolised as first-attempt adjunct; technique standardised (straight-to-cuff bend) but operator skill and intubation conditions varied as in real practice.
  • Separation of the variable of interest: First-attempt success 78.2% vs 71.5%; second attempt required 17.8% vs 26.1% (number of laryngoscopy attempts distribution differed, consistent with separation).
  • Adjunctive therapy use: Rescue adjuncts were permitted after first-attempt failure; post-failure use of a stylet occurred more often in the stylet group (66/109; 60.6%) than control (9/142; 6.3%), limiting contamination of the first-attempt comparison while allowing pragmatic rescue.
  • Outcome assessment: Primary endpoint (FPS) is objective; complication endpoints included objective components (e.g., SpO2 thresholds, cardiac arrest) and more subjective components (e.g., aspiration) with potential misclassification.
  • Statistical rigour: Primary endpoint powered and met; many secondary/exploratory outcomes used multiplicity adjustment (false discovery rate), increasing protection against false positives at the cost of power for secondary signals.

Conclusion on Internal Validity: Overall, internal validity appears moderate-to-strong, supported by concealed randomisation, balanced attrition, objective primary endpoint, and high protocol adherence; limitations chiefly relate to unavoidable operator unblinding and the mixed objectivity of complication outcomes.

External Validity

  • Population representativeness: Adult ICU patients requiring urgent intubation for invasive ventilation; common ICU indications (acute respiratory failure ~49% vs ~47%; coma ~25%; shock ~5%).
  • Important exclusions: Cardiac arrest at the time of intubation was excluded, so findings should not be extrapolated to peri-arrest intubations.
  • Device/technique context: The tested strategy is specific to direct laryngoscopy with a standard stylet configuration; applicability to routine video laryngoscopy practice is uncertain.
  • System-level applicability: Conducted in 32 French ICUs with an established intubation bundle culture and high RSI uptake; benefit and safety may differ where training, staffing, or bundle adherence differ.

Conclusion on External Validity: Generalisability is good to ICUs using direct laryngoscopy with RSI-based practice and similar staffing, but more limited where video laryngoscopy is standard, in peri-arrest intubations, or in systems without comparable team/bundle support.

Strengths & Limitations

  • Strengths:
    • Large, multicentre ICU RCT (32 ICUs) with a clinically relevant procedural primary endpoint.
    • Pragmatic conduct aligned with ICU “bundle” practice, enhancing real-world relevance.
    • Clear intervention standardisation (stylet configuration) and objective primary outcome.
    • Pre-specified protocol and statistical framework published prior to completion. 1
  • Limitations:
    • Unavoidable operator unblinding with potential bias for subjective outcomes (e.g., aspiration; perceived difficulty).
    • Primary endpoint is a surrogate for patient-centred outcomes; clinical benefit depends on whether improved FPS reduces harms.
    • Complication outcomes are common but heterogeneous; some components may be influenced by co-interventions and local practice.
    • Direct laryngoscopy-only context may be less aligned with contemporary increases in video laryngoscopy uptake.
    • Rare harms (major airway injury) remain imprecisely estimated despite large sample size.

Interpretation & Why It Matters

  • Clinical signal
    Routine first-attempt stylet use improved FPS (absolute +6.7%), but did not reduce the composite of early intubation-related complications or mortality.
  • Practical implication
    In ICUs where direct laryngoscopy remains common, a stylet-first approach is a reasonable default for increasing FPS, provided teams maintain vigilance for complications and retain a structured rescue plan.
  • What it refines
    Improving technical success (FPS) is necessary but not sufficient to reduce complications; haemodynamic optimisation, preoxygenation, and team/process factors likely remain dominant determinants of outcome.

Controversies & Subsequent Evidence

  • Surrogate endpoint versus patient-centred benefit: The trial demonstrated improved FPS but no reduction in early complications; commentary highlighted that “first-pass success without adverse events” may better reflect net clinical benefit than FPS alone. 2
  • Adjunct choice (stylet vs bougie) remains context-dependent: In critically ill adults, a subsequent ICU/ED RCT comparing bougie-first vs tube+stylet-first found no difference in FPS, contrasting with an ED RCT in difficult airways where bougie improved FPS vs tube+stylet. 34
  • Shifting device landscape: Subsequent large pragmatic evidence suggests video laryngoscopy increases FPS in critically ill adults versus direct laryngoscopy, and updated meta-analyses support improved first-pass success and fewer oesophageal intubations with video laryngoscopy; how routine stylet use interacts with video laryngoscopy in ICU remains less directly tested. 56
  • Process optimisation may dominate complication risk: Contemporary trials and syntheses emphasise preoxygenation strategy (e.g., non-invasive ventilation) as a key modifiable factor for peri-intubation hypoxaemia, potentially exerting larger effects on complications than small technical adjunct effects alone. 78
  • Guidelines and consensus statements: ICU intubation guidance continues to prioritise oxygenation, minimising attempts, and structured rescue planning, with adjuncts (stylet/bougie) framed as tools rather than universal mandates; broader difficult-airway and safety guidelines reinforce capnography-based confirmation and systems approaches to prevent catastrophic misplacement. 910111213

Summary

  • In 32 French ICUs, routine first-attempt stylet use increased first-pass intubation success (78.2% vs 71.5%; RR 1.10; 95% CI 1.02 to 1.18).
  • The improved first-pass success did not translate into fewer intubation-related complications within 1 hour (38.7% vs 40.2%; RR 0.96; 95% CI 0.83 to 1.12).
  • Severe complications were frequent (~26%) and were unaffected by stylet use; traumatic injuries were uncommon (~4%) and similar between groups.
  • No differences were observed in exploratory clinical outcomes including ICU-free days, ventilator-free days, or 28- and 90-day mortality.
  • The results support a stylet-first approach to improve technical success with direct laryngoscopy, while reinforcing that broader physiological and team/process strategies remain essential to reduce complications.

Overall Takeaway

STYLETO provides robust ICU RCT evidence that routine first-attempt stylet use during direct laryngoscopy modestly increases first-pass success, with no detectable reduction in early complications or mortality. The trial supports stylet-first as a reasonable technical default where direct laryngoscopy is common, while underscoring that complication prevention in ICU intubation still hinges on physiological optimisation, limiting attempts, and structured team processes.

Overall Summary

  • Routine stylet use improved first-pass success (absolute +6.7%) but did not reduce early complications.
  • Severe complications remained common (~26%), reinforcing that physiology/process factors dominate peri-intubation risk.
  • Best applied as part of a broader ICU intubation strategy (oxygenation, haemodynamic preparation, and early rescue planning).

Bibliography