Neuromuscular blockers in early ARDS

ACURASYS Study Investigators.NEJM;2010;363:1107-1116.doi:10.1056/NEJMoa1005372

Clinical Question

  • In patients with moderate-severe ARDS does the early use of a neuromuscular blocking agent improve mortality?


  • Randomised controlled trial
  • Computer-generated random-number tables with telephone system to ensure allocation concealment
  • Block randomisation
  • Double blinded
  • Sample size calculation: assuming 50% mortality, 340 patients required to detect 15% absolute reduction in mortality
    • 20% false negative rate
    • 5% false positive rate


  • 20 ICUs, France
  • March 2006 – March 2008


  • Inclusion: adult patients with moderate/severe ARDS as defined by:
    • endotracheal mechanical ventilation for acute hypoxaemic respiratory failure
    • and the presence of all of the following for <48 hours
      • PaO2/FiO2 <150 with PEEP ≥5, Tidal volume 6-8ml/kg
      • Bilateral pulmonary infiltrates and absence of clinical evidence of left atrial hypertension
  • Exclusion: continuous infusion of neuromuscular blocking agent at enrolment, pregnancy, raised ICP, longterm O2 therapy or home ventilation, severe chronic liver disease, bone marrow transplant or chemotherapy induced neutropenia, pneumothorax, expected duration of mechanical ventilation <48 hours or decision to withhold life sustaining treatment, actual body weight >1kg/cm height
  • 339 patients included in analysis


  • Neuromuscular blocking agent for 48 hours
    • cisatracurium 15mg bolus followed by infusion of 37.5mg/hr for 48 hours


  • Placebo

All patients

  • treated with volume assist-control mode of ventilation, tidal volumes of 6-8ml/kg of predicted body weight. Target SpO2 88-95% or PaO2 55-80mmHg. FiO2 and PEEP adjusted according to ARMA trial protocol.
  • allowed two open-label boluses of 20mg cisatracurium  if plateau pressue >32cmH20 despite increasing sedatives and decreasing tidal volume/PEEP
  • Use of peripheral nerve stimulators was not permitted
Comparing baseline characteristics of cisatracurium vs. placebo group
  • No significant difference in
    • tidal volume (6.55ml/kg vs. 6.48ml/kg)
    • PEEP (9.2 vs 9.2)
    • use of prone position, inhaled nitric oxide or almitirine mesylate (18.6% vs. 14.2%)
    • direct lung injury (80.2% vs. 75.9%)
    • main reason for ICU admission
      • medical (72.9% vs. 69.8%)
    • SAPS II (50 vs 47)
  • Patients in cisatracurium group had significantly lower
    • Pao2:FiO2 (106 vs. 115, p=0.03)
Comparing co-interventions during study period of cisatracurium vs. placebo group
  • No significant difference in use of
    • adjunctive therapies
      • prone position (28% vs. 29%)
      • inhaled nitric oxide (28% vs. 33%)
      • almitirine bismesylate (3% vs. 6%)
    • corticosteroids for ARDS (16% vs. 23%, p=0.10)


  • Primary outcome: adjusted 90 day mortality
    • significantly lower in cisatracurium group
    • Adjusted Hazard ratio 0.68 (95% C.I. 0.48-0.98), p=0.04
  • Secondary outcomes:
    • actual 90 day mortality – no significant difference
      • 31.6% vs. 40.7%, p=0.08
    • Comparing cisatracurium group vs. placebo
      • significantly more patients in placebo group had
        • barotrauma
        • pneumothraces
        • received open-label cistracurium in 1st 48 hours following inclusion
      • significantly more patients in cisatracurium group had
        • more days without non-respiratory organ failure
    • No significant difference in incidence of ICU-acquired paresis
  • Sub-group analysis based on PaO2:FiO2 ratio
    • probability of survival to day 90 (log-rank test)
      • PaO2:FiO2 <120, p=0.051
      • PaO2:FiO2 ≥120, p=0.74

Authors’ Conclusions

  • The early use of a neuromuscular blocking agent in severe ARDS may improve outcomes.


  • Randomised controlled trial
  • Double blinded (Clinicians likely to have been able to determine clinically if patient had received neuromuscular blocking agents and therefore may have treated patients differently)
  • Multi-centre
  • Protocolised ventilation strategy


  • With lower than expected mortality the power of the study is much lower than predicted.
  • The use of adjusted mortality for primary outcome was not specified when registering the trial on clinical trials
  • 10% of patients excluded due to ‘other reason’
  • Some aspects of management are not comparable to current UK practice (21% swanz-ganz, 30% nitric oxide)
  • Trend towards significant difference in use of corticosteroids

The Bottom Line

  • This multi-centre randomised controlled trial found a significant improvement in mortality for patients with severe ARDS who were treated with early neuromuscular blocking agents. Further studies to confirm this finding would be beneficial. These trials should ideally have standardised protocols for the use of co-interventions such as proning and the use of steroids. In the mean time I will need a good reason to not use neuromuscular blocking agents in early severe ARDS.

One comment

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.