Nasal High-Flow Therapy during Neonatal Endotracheal Intubation

@katehodg18. NEJM 2022;386:1627-1637; DOI: 10.1056/NEJMoa2116735

Clinical Question

• In neonates undergoing elective endotracheal intubation, does nasal high-flow therapy compared with standard care, improve 1st pass success without physiological instability?

Background

• A multicentre, international registry reported that 1st pass success in neonatal intubation was only 50%
• Neonates have a lower functional residual capacity and a greater metabolic demand than older children and adults. This can contribute to physiological instability during intubation attempts. The same registry reported that in ~half of the intubations performed, the neonates SpO2 dropped by ≥20%
• High flow nasal oxygen has been used in adults to aid intubation and prolong apnoea time

Design

• Randomised controlled trial
  • Computer-generated random-assignment sequence based on permuted blocks with varying block sizes
  • Stratified according to trial centre, post-menstrual age (≤28 or >28 weeks), and the use of premedication for intubation
• Non-blinded
• Intubation attempt defined as insertion of laryngoscope blade beyond the lips until its removal from the mouth
• Physiological instability defined as decrease in SpO2 of 20% from baseline or HR <100
• All intubations were recorded, including recording of oxygen saturation monitor
• Primary outcome assessed by independent assessment of video
• Sample size calculation
  • 246 intubations would give 90% power to detect an increase in the primary outcome from 30% to 50% with a 2 sided alpha level of 0.05
• Modified intention to treat analysis (excluded patients that had exclusion criteria)
• Primary outcome adjusted for randomisation stratification factors
• Registered on Australian New Zealand Clinical Trials Registry

Setting

• Two tertiary neonatal intensive care units in Melbourne, Australia
• Data collected: 2018 – 2021

Population

• Inclusion criteria:
  • Infants undergoing oral endotracheal intubation in the delivery room or neonatal intensive care unit
• Exclusion:
  • Nasal intubation
  • Urgent intubation as determined by the treating clinician
  • HR <120 beats
  • Contraindication to high-flow therapy (e.g., congenital nasal anomaly, congenital diaphragmatic hernia, or abdominal wall defect)
  • Cyanotic congenital heart disease
  • Suspected or proven severe COVID infection in the infant or mother
• 462 intubations were eligible for randomisation. 258 intubations randomised. 7 excluded post randomisation
• Comparing baseline characteristics of intervention vs. control group
  • Median gestational age: 27 vs 27 weeks
  • Median birth weight: 893 vs 841g
  • Median Apgar score at 5 min: 8 vs 7.5
  • Median age at intubation: 7 vs 13 hours
  • Median weight at intubation: 976g vs 907g
  • Location of intubation
    • Delivery room: 25% vs 27%
    • Neonatal intensive care unit: 75% vs 73%
  • FiO2 before intubation: 0.62 vs 0.62
  • Primary indication for intubation
    • Hypoxia: 59% vs 58%
    • Apnoea: 21% vs 20%
    • Resuscitation: 2.4% vs 5.5%
  • Operator experience
    • <20 previous intubations: 49% vs 40%
    • >20 previous intubations: 51% vs 60%
  • Use of videolaryngoscopy
    • 9.7% vs 7.1%

Intervention

• High flow nasal oxygen
  • Applied immediately before laryngoscopy, set at 8l/min
  • FiO2 was set at the concentration delivered before laryngoscopy and was increased to 100% oxygen if SpO2 <90%
  • Discontinued after 1st intubation attempt

Control

• Standard care
  • Laryngoscopy performed without high-flow nasal oxygen or supplemental oxygen

Management common to both groups

• The pre-intubation FiO2, the use of VL and the duration of intubation attempt was at the discretion of the treating clinician
• Pre-medication with atropine, fentanyl and suxamethonium was standard practice for intubations that occurred outside the delivery room

Outcome

• Primary outcome: Successful intubation on the 1st attempt without physiological instability – significantly greater in the intervention group
  • 50% vs 31.5%
  • Adjusted risk difference 17.6% (95% CI 6-29.2)
  • NNT 6 (95% CI 4-17)
• Primary outcome component analysis:
• Comparing intervention vs. control group
  • Significantly greater in intervention group
    • Successful intubation on 1st attempt
      • 68.5% vs 54.3%
      • Adjusted risk difference: 15.8 (95% CI 4.3-27.3)
    • No physiological instability
      • 63.7% vs 50.4%
      • Adjusted risk difference: 13.4% (95% CI 1.3-25.5)
• Secondary outcomes:
• Comparing intervention vs. control group
  • Median SpO2
    • 94% vs 89%
    • Difference: 5 (95% CI 1.1-8.9)
    • Time to desaturation
      • 44.3 vs 35.5s
  • Significantly less in intervention group
    • Intubations in which desaturation occurred
      • 28.2% vs 39.4%
  • No significant difference in
    • Median duration of 1st intubation attempt
      • 124 vs 127 seconds
    • Oesophageal intubation
      • 14.5% vs 15.7%
  • Median time to apply nasal prongs in intervention group: 9.9s
• Adverse events, comparing intervention vs control group
  • CPR or adrenaline within 1 hour of intubation attempt
    • 0% vs 1.6%
  • Pneumothorax within 72 hours post randomisation
    • 1.6% vs 4.7%
  • Death within 72 hours post randomisation
    • 0.8% vs 2.4%
• Sub-group analysis for primary outcome, comparing intervention vs standard group
  • Operator experience
    • <20 intubations: 49.2% vs 15.7%
      • Adjusted risk difference 33.3% (95% CI 18.3-48.2)
    • ≥20 intubations: 50.8% vs 42.1%
      • Adjusted risk difference 7.5% (-9.4 to 24.3)

Authors’ Conclusions

• In neonates the use of high-flow therapy during oral endotracheal intubation led to greater likelihood of successful intubation without physiological instability

Strengths

• Randomised controlled trial
• Multi-centre
• Video recording of intubation attempt and SpO2 monitor, with independent review of video to assess primary outcome
• Stratified according to operator experience

Weaknesses

• Clinicians were not blinded but impossible to do, and use of sham high flow may have affected standard care as may have prolonged intubation attempt
• Only conducted in 2 tertiary centres in 1 country, limiting external validity

The Bottom Line

• For neonates undergoing intubation, the first pass success rate without physiological instability is low
• The application of high flow nasal oxygen immediately prior to laryngoscopy significantly increased the rate of successful intubation on the 1st attempt without physiological instability – this is particularly pronounced for operators with limited experience

External Links

• [article] Nasal High-Flow Therapy during Neonatal Endotracheal Intubation
• [further reading] Neonatal intubation practice and outcomes: an international registry study.
• [further reading] Transcutaneous humidified rapid insufflation ventilatory exchange for oxygenation of children during apnoea: A prospective randomised controlled trial

Metadata

Summary author: David Slessor
Summary date: 28th April 2022
Peer-review editor:Segun Olusanya

Image by: Image by RitaE from Pixabay