Nocturnal Dexmedetomidine for Delirium Prevention

From Wiki Journal Club
Jump to navigation Jump to search
Skrobik Y. "Low-Dose Nocturnal Dexmedetomidine Prevents ICU Delirium. A Randomized, Placebo-controlled Trial.". Am J Respir Crit Care Med. 2018. 197(9):1147-1156.

Clinical Question

In critically ill adults requiring sedation and analgesia, does the addition of nocturnal dexmedetomidine to standard care decrease the risk of developing delirium during ICU stay?

Bottom Line

Nocturnal low-dose dexmedetomidine appears to decrease the rate of delirium in critically ill adult patients that require sedation.

Major Points

The symptom of delirium is an independent predictor of worse patient outcomes, including increase length of stay, costs, mortality, and is associated with long term morbidity associated with cognitive decline. [1] The cause of delirium is still debated so as yet an effective prophylaxis has yet to be identified. Sleep disruption due to critical illness and medical intervention is suggested as a cause/contributing factor in delirium.

This trial, published in 2018, included 100 patients from two centres (Canada and America) the randomized patient to nocturnal sedation with low-dose dexmedetomidine to maintain a Richmond Agitation and Sedation Score of -1 or placebo. Patients in the intervention arm provided an ARR of 26% for patients to be delirium free (95% CI 0.23–0.82; P = 0.006). Patient reported sleep quality was similar between the two groups as well as adverse effects.

The SPICE III published in 2019


As of June 2018, no guidelines have been published that reflect the results of this trial.


  • Two-centre, prospective, double-blind, placebo-controlled trial
  • N=100
    • Dexmedetomidine (n=50)
    • Placebo (n=50)
  • Setting: a Canadian medical-surgical ICU and an American medical ICU
  • Enrollment: 24Feb2013 to 22Jan2016
  • Mean follow-up:
  • Analysis: Intention-to-treat
  • Primary Outcome: Proportion delirium free during critical illness


Inclusion Criteria

  • Admission to ICU
  • Requiring intermittent or continuous sedation
  • expected to require ≥48 hours of ICU care

Exclusion Criteria

  • Delirium prior to randomization (ICDSC score ≥ 4)
  • history fo severe dementia
  • conditions precluding delirium assessment (ie Traumatic Brain Injury)
  • symptoms precluding dexmedetomidine administration (ie bradycardia)
  • previous exposure to study drug or clonidine
  • pregnancy

Baseline Characteristics

‘'Dexmedetomidine Group displayed

  • Demographics: mean age 62 years, 62% male
  • Physiologic parameters: mean APACHE-II score 23.6, mechanically ventilated 90%,
  • Anthropomorphics: Weight
  • Admission diagnosis: Sepsis/ARDS 18%, pneumonia 18%, major surgery 18%, Respiratory failure 16%, Trauma 10%, Gastrointestinal 4%, Cardiac 4%, Other 12%
  • Comorbidities: Hypertention 54%, Smoking 30%, ≥2 alcoholic drinks/day 14%, Drug misuse 2%, Depression 0%, heart failure 2%, COPD 22%, neurodegenerative disease 2%, Renal failure 6%, Liver disease 2%, Cancer 10%


  • during nocturnal period of 21h30 to 06h30, regular sedatives were halved and dexmedetomidine or placebo infused
    • dexemedetomidine 0.2 mcg/kg/h of ideal body weight titrated to maintain RASS -1 to a maximum of 0.7 mcg/kg/h.
  • As needed sedation was given as midazolam 1-5 mg intravenously every 1 hour as needed


Comparisons are Dexmedetomidine vs. Placebo.

Primary Outcomes

Free of delirium during ICU admission
80% vs. 54% (ARR 26%, 95% CI 0.23-0.82, P = ) NNT4

Secondary Outcomes

Median ICU days free of delirium
8 days vs. 6 days (P < 0.001)
Median duration of first delirium episode
2.0 days vs. 2.2 days (P = 0.73)

Subgroup Analysis

Adverse Events


  • Sedatives in placebo arm was also halved during the nocturnal period, potentially leading to unblinding or bias towards intervention
  • all other sedation and analgesia was at the treating clinician discretion
  • at enrolment, fewer patients in the intervention arm had a RASS score of -2 or -3 (8% vs. 28%, P = 0.008) which may have led to bias


Further Reading