SLEAP: Daily sedation interruption in mechanically ventilated critically ill patients cared for with a sedation protocol: a randomized controlled trial

Mehta et al for the SLEAP Investigators. JAMA 2012; 308(19):1985-92.

Clinical Question

  • In patients who are mechanically ventilated and sedated with opioids and / or benzodiazepines, do daily sedative interruptions reduce the duration until extubation?


  • Randomised, controlled trial
  • 65-patient 3-centre pilot trial conducted previously
  • Block randomisation according to centre
  • Allocated by automated telephone system
  • No blinding attempted


  • 14 Canadian and 2 US intensive care units
  • January 2008 – July 2011


  • Inclusion: expected to require at least 48 hours mechanical ventilation; ICU team had initiated continuous sedative / opioid infusion.
  • Exclusion: cardiac arrest, traumatic brain injury, administration of neuromuscular blocking drugs.
  • 2091 screened of which 430 randomised


  • Opioid and benzodiazepine infusions were stopped daily.
  • They were restarted at half the previous rate when deemed necessary by the bedside clinician.
    • Mean SAS score was 3.28 (95% CI 2.92–3.85)


  • No planned daily interruption
    • Mean SAS score was 3.23 (95% CI 3.0–3.71)
  • Both groups were sedated by opioids (morphine or fentanyl) for pain and benzodiazepines (midazolam or lorazepam) for agitation.
  • Hourly assessment by either Richmond Agitation Sedation Scale or the Sedation–Agitation Scale directed the titration either up or down of these drugs.
  • Target SAS score was 3 (sedated but rousable) to 4 (calm and cooperative) – “light” sedation
    • Comparable RASS score is -3 to 0.
  • The bedside nurse chose whether to titrate opioids or benzodiazepines.
  • Propofol, ketamine and dexmedetomidine were not allowed.


  • Primary outcome: there was no difference in median time to successful extubation.
    • median days to extubation 7 days in both groups (HR 1.08, 95% CI 0.86–1.35; p=0.52)
  • Secondary outcome:
    • a priori sub-group analysis demonstrated a reduced median duration to extubation in trauma and surgical patients, and no difference in medical patients.
    • Nurse workload was significantly higher in the intervention group.
    • There were no differences in accidental event rates

Authors’ Conclusions

  • In critically ill patients who are mechanically ventilated, when nurses implement a sedation protocol that targets light sedation, the use of daily sedative interruptions does not reduce the duration until extubation and may increase both the amount of sedative administered and the nursing workload.


  • Just 7 lost-to-follow-up due to withdrawal of consent. Unlikely to affect result significantly.
  • Pragmatic design: did not use additional research nurses at bedside.
  • Multi-centre and mixed patient cohorts (surgical and medical) increases generalisability.


  • Physician refusal to enrol 10.8% of screened patients. Selection bias?
  • Re-intubation rate 5.6% and 7.7%; Tracheostomy rate 23.2% and 26.3%. Perhaps higher than UK practice? Might reduce external validity (generalisability).
  • Adherence was poor:
    • Daily interruption were administered on 85.6% of all eligible patient-days.
    • 53% of patients missed at least 1 daily interruption and 6 missed every daily interruption!
    • Infusions were restarted at higher than 50% for 14% of eligible patients.
    • Propofol use occurred on 3% and 2% of study days for intervention and control groups respectively.
    • Interruptions occurred in 16% of control patients, accounting for 2.3% of study days.
    • These are likely to reduce any true difference and will bias toward a more neutral result. They emphasise the difficulty in implementing sedation breaks consistently.
  • Did not allow Propofol, which is widely used for sedation on ICU. Benzodiazepines used less widely now? Reduces generalisability.

The Bottom Line

  • If a sedation protocol is used to target “light” sedation within an ICU, daily interruptions are not shown to be beneficial, are very hard to implement consistently and may increase nursing workload.


Full text pdf / abstract / doi: 10.1001/jama.2012.13872

Editorial, Commentaries or Blogs


Summary author: @DuncanChambler
Summary date: 10th May 2014
Peer-review editor: @stevemathieu75

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.