From Wiki Journal Club
Jump to: navigation, search
Ferguson ND, et al. "High-Frequency Oscillation in Early Acute Respiratory Distress Syndrome". The New England Journal of Medicine. 2013. 368(9):795-805.
PubMedFull textPDF

Clinical Question

In patients is moderate to severe early ARDS, does high-frequency oscillatory ventilation reduce in-hospital mortality?

Bottom Line

In patients with moderate to severe early ARDS, early high-frequency oscillatory ventilation might increase in-hospital mortality

Major Points

The high mortality rates of ARDS have been improved upon by the ARDSNet trial's (2000) protocol of low VT (6 mL/kg) and has become standard of care in ICUs. Development of high-frequency oscillatory ventilation (HFOV) was thought to further improve survival through prevention of atelectrauma given its ability to deliver even smaller VT of 1-2 mL/kg delivered at 3-15 breaths per second.[1] Early trials on HFOV like MOAT[2] (2002) indicated a potential benefit for the therapy, though interpretation of the results was muddled by lack of comparison of HFOV to a modern, low VT strategy.

The 2013 OSCILLATE trial randomized 548 intubated ICU patients with early moderate-to-severe ARDS to HFOV or a ventilatory mode similar to that used in ARDSNet with a primary outcome of in-hospital mortality. The trial was stopped early because of the interim analysis demonstrating a 12% absolute increase in in-hospital mortality with HFOV (47% vs. 35%, NNH 8). The similar OSCAR trial[3] (2013) published concurrently explored the use of HFOV in all patients with ARDS. It did not demonstrate a mortality benefit.


Surviving Sepsis Campaign severe sepsis and septic shock (2016, adapted)[4]

  • Recommend against using high-frequency oscillatory ventilation in adults with ARDS from sepsis (strong recommendation, moderate quality of evidence)


  • Multicenter, prospective, non-blinded, randomized controlled trial
  • N=548 intubated patients with early, moderate-to-severe ARDS
    • HFOV (n=275)
    • Control (n=273)
  • Setting: 39 ICUs in 5 countries
  • Enrollment: 2009-1012 (stopped early given significant findings)
  • Follow-up: Until hospital discharge
  • Analysis: Intention-to-treat
  • Primary outcome: In-hospital mortality


Inclusion Criteria

  • Intubated patients with pulmonary symptoms for ≤2 weeks
  • ARDS defined by:
    • Hypoxemia with PaO2:FIO2≤200 when FIO2≥0.5
    • Bilateral airspace disease

Exclusion Criteria

  • Left atrial hypertension causing hypoxemia
  • Pulmonary hemorrhage from vasculitis
  • Neuromuscular disorders likely to prolong need for ventilation
  • "Severe chronic respiratory disease"
  • Expected 6 month mortality from preexisting illness
  • At risk for intracranial hypertension
  • "Lack of commitment to life support"
  • Expected time on ventilator <48 hours
  • Age <16 or >85 years
  • Weight <35 kg or >1kg/cm of height
  • Meeting of eligibility criteria for >72 hours
  • Already receiving HFOV
  • Physicians declining enrollment

Baseline Characteristics

From the HFOV group.

  • Demographics: Age 55 years, female 39%
  • APACHE II score: 29
  • ARDS risk factors:
    • Sepsis: 47%
    • Pneumonia: 56%
    • Gastric aspiration: 18%
    • Trauma: 4%
    • Other: 26%
  • VT: 7.2 mL/kg
  • Plateau pressure: 29 cmH2O
  • PEEP: 13 cmH2O
  • Minute ventilation: 11.3 L/min
  • Oxygenation index: 19.6
  • PaO2:FIO2: 121 mmHg
  • PaCO2: 46 mmHg
  • Arterial pH: 7.32
  • Barotrauma: 7%
  • Other interventions:
    • Vasopressors: 67%
    • RRT: 11%
    • Glucocorticoids: 34%
    • Neuromuscular blockers: 31%


  • Patients enrolled and were maintained on pressure-control mode with VT 6 mL/kg and FIO2 of 0.60, and PEEP ≥10 cmH2O as needed
  • If PaO2:FIO2 was ≤200, patients were randomized to one of two groups:
    • HFOV:
      • Recruitment of 40 cmH2O for 40 seconds
      • Mean airway pressure of 30 cmH2O with pressure adjusted to maintain a PaO2 of 55-80 mmHg
      • VT minimized to keep pH >7.25
      • Reversion to standard ventilation of mean airway pressure was ≤24 cmH2O for 12 hours or anytime when ≤20 cmH2O
      • Re-initiation of HFOV if FIO2 >0.4 or PEEP >14 cmH2O for >1 hour in the subsequent 48 hours
    • Control:
      • Recruitment of 40 cmH2O for 40 seconds
      • Pressure control mode, VT of 6 mL/kg with plateau pressures ≤35 cmH2O, and PEEP of 20 cmH2O with adjustments made per the protocol
      • Patients could be maintained on pressure-support or volume-assist modes
      • Weaning protocol using spontaneous breathing trials
  • Both groups could reduce PEEP or mean airway pressure based upon clinical judgement of lung overdistension
  • If requiring FIO2 ≥0.9, hypoxemia protocols like NO or change in positioning could be instituted
  • Any alternative therapy could be started for three groups:
    • Refractory hypoxemia - defined as PaO2 <60 mmHg for ≥1 hour with FIO2 of 1.0 and neuromuscular blockade
    • Refractory barotrauma - defined as persistent pneumothorax or worsening subcutaneous emphysema despite two thoracostomy tubes on the affected side
    • Refractory acidosis - pH ≤7.05 despite neuromuscular blockade
  • Medical management was otherwise left to the attending physicians


Presented as HFOV vs. control.

Primary Outcome

In-hospital mortality
47% vs. 35% (RR 1.33; 95% CI 1.09-1.64; P=0.005; NNH=8)

Secondary Outcomes

ICU mortality
45% vs. 31% (RR 1.45; 95% CI 1.17-1.81; P=0.001)
28 day mortality
40% vs 29% (RR 1.41; 95% CI 1.12-1.79; P=0.004)
New barotrauma
18% vs. 13% (RR 1.37; 95% CI 0.91-2.06; P=0.13)
Refractory barotrauma: <1% vs. <1% (NS)
New tracheostomy
22% vs. 25% (RR 0.87; 95% CI 0.64-1.19; P=0.39)
Refractory hypoxemia
7% vs. 14% (RR 0.50; 95% CI 0.29-0.84; P=0.007)
Refractory hypoxemia mortality: 79% vs. 66% (RR 1.20; 95% CI 0.87-1.66; P=0.31)
Refractory acidosis
3% vs. 3% (RR 1.20; 95% CI 0.44-2.85; P=0.82)
Mechanical ventilation duration in survivors
11 days vs. 10 days (NS)
ICU duration in survivors
15 days vs. 14 days (NS)
Hospitalization duration in survivors
30 days vs 25 days (NS)


  • The authors identify that the trial was stopped early despite the stopping thresholds not being met and that the high mean airway pressure HFOV protocol may have led to worse outcomes than one with lower pressures
  • The sedation strategy in the HFOV group may have led to higher mortality as patients may have been given higher doses of sedative and, subsequently, fluids[5]
  • The HFOV may have been initiated too late[5]
  • Successful HFOV may take expertise not present at the study centers[5]


  • Canadian Institutes of Health
  • King Abdullah International Medical Research Center
  • Authors with financial ties to private groups

Further Reading

  1. Malhotra A and Drazen JM. "High-Frequency Oscillatory Ventilation on Shaky Ground." The New England Journal of Medicine. 2013;368:863-865.
  2. Derdak S et al. "High-Frequency Oscillatory Ventilation for Acute Respiratory Distress Syndrome in Adults A Randomized, Controlled Trial." American Journal of Respiratory and Critical Care Medicine (2002):166:801-808.
  3. Young D et al. "High-frequency oscillation for acute respiratory distress syndrome." The New England Journal of Medicine. 2013;368:806-813.
  4. Rhodes A, et al. "Surviving Sepsis Campaign: International guidelines for management of sepsis and septic shock: 2016." Critical Care Medicine. 2017;45(3)1-67.
  5. 5.0 5.1 5.2 Multiple authors. "Correspondence: High-frequency oscillation for ARDS." The New England Journal of Medicine. 2013;368(23):2231-2234