Randomised Controlled Trial of Humidified High-Flow Nasal Oxygen for Acute Respiratory Distress in the Emergency Department

Jones. Respiratory Care 2016; 61(3):291-299. doi:10.4187/respcare.04252

Clinical Question

  • In hypoxic adults presenting to the emergency department, does nasal high-flow oxygen compared to standard oxygen therapy reduce the need for advanced respiratory support?


  • Single-centre, randomised, controlled trial
  • Randomised by custom Microsoft Access application
  • Sequence concealed by sealed, opaque envelopes
  • Data collected from clinical notes by non-blinded study assessor
  • Intention to treat analysis
  • 80% power at α=0.05 if 390 patients recruited
    • Baseline conversion to advanced respiratory support of 24% based upon pilot observational study
    • Expected relative reduction of 33% with nasal high-flow oxygen


  • Single tertiary academic city hospital in New Zealand
  • July 2012 to May 2014


  • Inclusion: Patient presenting to hospital via the emergency department with
    • Arrival or pre-hospital SpO2 < 92% (<90% if known chronic CO2 retention)
    • And respiratory rate > 22 breaths/minute
  • Exclusion: Intubated pre-hospital or received non-invasive ventilation (NIV) immediately upon arrival; bullous lung disease; pneumothorax; facial abnormalities including trauma or trans-nasal neurosurgery; epistaxis within last 2 weeks
  • 1,287 patients screened, 322 were randomised, 303 were analysed (19 excluded from analysis)


  • High-flow nasal oxygen
    • Airvo1 or Airvo2 (Fisher & Paykel Healthcare)
    • Starting 40 l/min at 37°C and FiO2 0.28
    • Titrated according to clinical need


  • Standard oxygen therapy
    • Hudson mask, Venturi device or nasal prongs
    • Administered at 1-15 l/min
    • Titrated according to clinical need

Management common to both groups

  • The decision to convert to non-invasive ventilation (NIV) was according to the British Thoracic Society’s guidelines
    • Persistent or worsening hypoxaemia
    • Worsening tachypnoea
    • Rising arterial PCO2
    • Persistent respiratory acidosis with pH < 7.30 despite 30 minutes of optimised standard acute medical therapy
    • Worsening fatigue or confusion
    • Decision by the treating clinician that escalation of therapy was necessary
  • Choice of NIV strategy (CPAP 5 cmH2O or BIPAP 5 and 10 cmH2O) was at clinician’s discretion
  • The decision to convert to invasive positive pressure ventilation (IPPV) was also according to the BTS guidelines
    • NIV criteria met and clinician anticipated that invasive ventilation will be required
    • Failure of NIV
    • Patient unable to maintain a patent airway or manage secretions


  • Primary outcome: no difference was seen in the rate of conversion to non-invasive ventilation (NIV) or invasive positive-pressure ventilation (IPPV)
    • HFNC: 3.6%
    • Standard O2: 7.3%
    • P = 0.20 by Fisher’s exact test
    • Absolute risk reduction: 3.6% (95% CI -1.57% to 8.79%)
    • Fragility Index: -3 patients
      • This means, if 3 patients who received NIV/IPPV in the HFNC group had the opposite outcome, this trial would then demonstrate a statistically significant difference
  • Secondary outcome:
    • Failure to tolerate therapy: clinically significant rate in HFNC group
      • HFNC: 8.5% (1 in 12 patients)
      • Standard O2: 0%
    • Emergency department length of stay: no difference
    • Hospital length of stay: no difference
    • 90 day mortality: no difference
      • HFNC: 21.2%
      • Standard O2: 17.4%
      • P = 0.47 by Fisher’s exact test
      • Absolute risk reduction: 3.82% (95% CI -5.06% to 12.7%)
    • Patients satisfaction survey: Overall, patients rated Standard O2 better than HFNC
      • However, 48% lost-to-follow-up
      • More patients described Standard O2 therapy as ‘comfortable’:
        • HFNC: 73.4%
        • Standard O2: 85.9%
        • P = 0.076

Authors’ Conclusions

  • Oxygen therapy delivered by high-flow nasal cannulae did not reduce the need for non-invasive or invasive ventilation when compared against standard oxygen therapy
  • However, the trial was unexpectedly underpowered and a true, clinically important benefit may exist, therefore this trial should be considered a pilot for a larger (~900 patient) trial
  • One in 12 patients are intolerant of oxygen delivery via high-flow nasal cannulae


  • Important theory given increasing use of nasal high-flow oxygen therapy
  • Sensible inclusion and exclusion list with good generalisability to most Emergency Departments
  • Randomised, controlled trial
  • Intention-to-treat analysis


  • Non-blinded with subjective primary outcome at clinician’s discretion
  • Failed to recruit target number, so under-powered to detect the expected or a smaller difference
  • Significant loss-to-follow-up for patient experience survey – strong likelihood of bias in reporting

The Bottom Line

  • This trial did not demonstrate a difference between nasal high-flow oxygen and standard oxygen therapy in hypoxic and dyspneic patients arriving in the Emergency Department
  • Significant weaknesses in the methodology mean that no firm conclusion can be drawn, as a false negative result may have occurred

External Links


Summary author: Duncan Chambler
Summary date: 5th July 2016
Peer-review editor: Steve Mathieu


Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.