Video Laryngoscopy vs Direct Laryngoscopy for Endotracheal Intubation in the Operating Room A Cluster Randomized Clinical Trial

Ruetzler et al. JAMA 2024;331;(15):1279-1286. doi:10.1001/jama.2024.0762

Clinical Question

• In patients being intubated for cardiac, thoracic or vascular surgery, does initial video laryngoscopy (VL) with a hyperangulated blade or direct laryngoscopy (DL) effect the number of intubation attempts?

Background

• Endotracheal intubation is integral to the safe provision of anaesthesia for many major surgical procedures
• Repeated intubation attempts are correlated with complications including hypoxia, haemodynamic instability, aspiration, airway trauma and cardiac arrest
• Video laryngoscopes were introduced in 2001 are increasingly replacing direct laryngoscopes as the preferred device for intubation especially in critically ill patients in ED and ICU settings
• VL may improve laryngeal visualization and result in improved first pass success
• VL vs DL has been examined in a number of large trials including:
  • DEVICE (NEJM 2023, n= 1420) demonstrated improved 1st pass success rate (85% in VL group vs 70% in DL) in a group of relatively inexperienced airway operators in ED and ICU settings with no difference in severe complications
  • MACMAN (JAMA 2017, n=371) did not demonstrate a difference in first pass intubation success in an ICU setting in both “expert” and non-expert airway operators. VL was associated with more frequent severe life threatening complications than DL (9.5% of patients in VL group vs 2.8% in DL)
• The most recent Difficult Airway Society guidelines (2017) for airway management in critically ill patients pre-date recent large trials and make no recommendation for VL vs DL

Design

• Cluster randomised multiple crossover clinical trial
  • Cardiac operating theatres split into 2 blocks of 11 theatres, each treated as a unit and randomised to VL or DL for 1 week blocks
    • ie half of the operating theatres used VL for 1 week  at a time whilst the other half used DL. Re-randomisation occurred at the end of the week allowing for multiple crossovers
  • Randomisation occurred in 1:1 ratio, unstratified, via web-based system
• Statistical analysis utilised modified intention-to-treat analysis
• Individual patient consent waived
• Unblinded clinicians

Setting

• Single academic hospital in USA (Cleveland Clinic)
• Enrollment from March 2021 to December 2022 (90 one-week blocks)
• Trial ceased early at 50% of planned recruitment for treatment effect favouring the VL group

Population

• Inclusion:
  • Adults >18 years old undergoing elective or emergent cardiac, thoracic or vascular surgery requiring single-lumen endotracheal intubation
• Exclusion:
  • Requiring awake fiberoptic intubation
  • Already intubated
  • Clinicians refused to participate
• n= 8429
• Baseline characteristics generally well matched
  • Mean age 63
  • Mean BMI 28
  • 65% male
  • 75% cardiac surgery
  • 70% of first attempts by nurse anaesthetist or resident
  • 80% ASA 4
• Reasons for exclusion generally well balanced

Intervention

• Hyperangulated VL
  • First intubation attempt with GlideScope video laryngoscope 3 or 4 hyperangulated blade with stylet

Control

• Direct laryngoscopy
  • First intubation attempt with standard geometry Macintosh blade

Management common to both groups

• All clinicians watched instructional video on use of hyperangulated VL and had oppourtunity to practice on a mannequin
• All intubations performed using hyperangulated VL for two weeks prior to trial commencement
• Minor external laryngeal manipulation (eg BURP) and cricoid pressure permitted (If appropriate, give details of patient care that was protocolised or common to both groups
• Up to 3 attempts allowed with initial device then clinicians could switch to any device they preferred
  • Intubation attempt defined as insertion of blade or endotracheal into mouth. Subsequent attempts defined by reinsertion of the ETT or a blade

Outcome

• Primary outcome: VL group had significantly lower number of intubation attempts vs DL group (estimated proportional odds ration 0.20 [95% CI 0.14-0.28, P< 0.001]).
  • ie patients in VL group had 80% higher odds of requiring fewer intubation attempts vs patients in DL group
  • Findings consistent across sensitivity analyses including staff refusal to participate in trial
  • VL group reported 98.3% 1st pass success rate vs 92.4% in DL group
• Secondary outcomes: Comparing VL vs. DL
  • Favouring VL
    • Lower rates of intubation failure (Switching to an alternative laryngoscopy device or not intubated after 3 attempts) 0.27% vs 4.0%, RR 0.06 95% CI 0.03-0.14 P<0.001
  • No significant difference in
    • Dental or airway injury – RR 0.87 (95% CI 0.48-1.58 P=0.53)

Authors’ Conclusions

• Hyperangulated video laryngoscopy decreased the number of attempts needed to achieve endotracheal intubation in the operating theatre

Strengths

• Largest trial of VL vs DL available to date by far
• Pragmatic approach to choice to induction agents, positioning and external manipulation
• Range of experience of airway operators reflects real-world practice
• High acuity population (80% of patients ASA 4 – “severe systemic disease that is a constant threat to life”) may increase external validity to ED and ICU populations

Weaknesses

• Use of a hyperangulated VL with stylet device decreases external validity
  • 3 variables at play which cannot be separated – blade geometry, use of stylet and video
  • These findings may not be generaliseable to “standard geometry” VL devices (eg CMAC standard blades or McGrath) or other hyperangulated blades (CMAC D-blade, McGrath X-blade) which can be used with a stylet or a bougie whereas GlideScope is typically used with the proprietary stylet
• Single center trial – reduces external validity
  • Manuscript offers no sense of usual practice prior to trial commencement
    • Pre-trial training only occurred for use of VL (implying that DL was standard practice) however there may have been clinicians unfamiliar with DL decreasing 1st pass success rates in the control
• Severe adverse events which are important to patients and clinicians (eg aspiration and death) were not assessed as secondary outcomes
• Number of intubation attempts as primary outcome is not patient centered
  • DEVICE trial found no association between number of intubation attempts and serious adverse outcomes although multiple smaller trials have found an association
• Ceased early for efficacy
  • May decrease the precision of the estimated treatment effect
  • May decrease sensitivity to detect rare but severe adverse effects eg death
  • Increased sample size may have allowed for greater subgroup analysis (eg by ASA, comorbidities, acute vs emergency surgery, airway operator experience) although this was not part of the statistical plan
• Video laryngoscopes were provided by manufacturers of GlideScope although no other commercial involvement is disclosed

The Bottom Line

• This large, well conducted single center RCT demonstrates the superiority of hyperangulated VL compared to DL for first pass success in a high acuity surgical population
• These findings may not be generalisable to other critical care environments especially where a different VL device is utilised
• Despite issues of external validity, these data reaffirm my practice of utilising a video laryngoscope, where available, as my default device for intubation

External Links

• article Video Laryngoscopy vs Direct Laryngoscopy for Endotracheal Intubation in the Operating Room A Cluster Randomized Clinical Trial

Metadata

Summary author: Daniel Chung
Summary date: 12/08/2025
Peer-review editor: David Slessor

Picture by: Terje Sollie/pexels