MIDEX-PRODEX

In Review

Jakob SM, et al. "Dexmedetomidine vs midazolam or propofol for sedation during prolonged mechanical ventilation: two randomized controlled trials". JAMA. 2012. 307(11):1151-1160.
PubMed • Full text

Clinical Question

In mechanically ventilated patients in the ICU, is dexmedetomidine non-inferior to midazolam or propofol in maintaining mild to moderate sedation?

Bottom Line

Dexmedetomidine was non-inferior to midazolam or propofol in maintaining target sedation levels and also offered the perquisite of improving the patient’s ability to communicate.

Major Points

Dexmedetomidine is a central alpha-2 agonist that is used to keep ventilated ICU patients sedated and has several attractive properties as compared to other medications used for this purpose such as benzodiazepines and propofol. Both are associated with respiratory depression, further benzodiazepines are associated with an increased risk of ICU related delirium and likely due to the lipid vehicle, propofol infusion syndrome (PRIS) is of concern with many patients.

These two parallel trials lined up two of the major sedative agents against dexmedetomidine: midazolam (MIDEX) and propofol (PRODEX). Using comparable designs and attempted to blind despite the obvious issues with the lipid vehicle with propofol. Demedetomidine was shown to meet the non-inferiority margin of 15% as compared to midazolam and propofol for time within trget sedation of RASS 0 to -3. Compared to the other agents, the dexemedtomidine group had shorter time to extubation, was able to communicate with the nursing staff more easily in terms of pain and cooperation, and was far more easily aroused. Despite these outcomes it did not show an difference in ICU or hospital length of stay.

The guidelines from ACCM in 2013 recommend dexmedetomidine to be used in select cases: when patients are experiencing delirium or at risk for developing delirium. This is likely due to the glaring discrepancy in costs between its use and other sedative options. Despite the faster time to extubation, the cost savings may not be realized given there was no difference in ICU or hospital length of stay.

Guidelines

This trial is reflected in the American College of Critical Care Medicine (ACCM) Guidelines for pain, agitation, and delirium in the ICU, Recommended dexmedetomidine use when:

• in patients at risk for developing delirium
• in patients that are experiencing delirium not associated with ethanol or benzodiazepine withdrawal

Design

• Phase 3, European, Multicenter, Randomized, Double Blind, control trial
• Enrollment: 2007-2010
• Follow-up: 45 days
• Analysis: Per-protocol
• Primary outcome: (1) Proportion of time in target sedation range without use of rescue therapy. (2) Duration of mechanical ventilation.

MIDEX

• N=501
  • Dexmedetomidine (n=249)
  • Midazolam (n=252)
• Setting: 44 centers in 9 European countries

PRODEX

• N=500
  • Dexmedetomidine (n=251)
  • Propofol (n=249)
• Setting: 31 centers in 6 European countries and Russia

Population

Inclusion Criteria

• 18 years of older
• Mechanically ventilated
• Requiring light to moderate sedation
• Use of midazolam or propofol expected to last >24 hours
• Randomized within 72 hours of ICU admission
• Randomized within 48 hours of starting continuous sedation

Exclusion Criteria

• Severe neurological disorder
• MAP < 55 mmHg
• Heart rate < 50 bpm
• Use of alpha-2 agonists or antagonists within 24 hours prior to randomization

Baseline Characteristics

MIDEX Dexmedetomidine group displayed

• Male: 61%
• Age, median(IQR), y: 65(55-74)
• SAPS II, median(IQR): 46(36-56)
• Main reason for admission to ICU
  • Medical: 73%
  • Surgical: 22%
  • Trauma: 5%
• Any infection at ICU admission: 58%
• Organ failures (SOFA score >2)
  • Respiratory: 60%
  • Cardiovascular: 61%
  • Renal: 15%
  • Coagulation: 8%
  • Liver: 2%
• Total SOFA score, median (IQR): 7 (5-9)

Interventions

• Study participants were kept sedated targeting a RASS score of 0 to -3.
• Dexmedetomidine 0.2 – 1.4 mcg/kg/h
  • MIDEX n=249
  • PRODEX n=251
• Midazolam (n=251) 0.03 – 0.2 mg/kg/h
• Propofol (n=247) 0.3 – 4mg/kg/h
• Pain treated with fentanyl as needed
• Daily sedation vacation and spontaneous breathing trials were used to assess for need for continued sedation
• Study drugs were continued for up to 14 days, stopped upon extubations

Outcomes

‘’Comparisons are dexmedetomidine vs. control drug (midazolam or propofol)’’

Primary Outcomes

Ratio of time at target sedation without rescue medication

MIDEX: 60.7% vs. 56.6% (1.07; 95% CI, 0.97–1.18; P=0.15)

PRODEX: 64.6% vs. 64.7% (1.0; 95% CI, 0.92–1.08; P=0.97)

Median duration of mechanical ventilation

MIDEX: 123 hours (67-337 hours) vs. 164 hours (IQR, 92-380 hours) [P=0.03]

PRODEX: 97 hours (45-257 hours) vs. 118 hours (IQR, 48-327 hours) [P=0.24]

Secondary Outcomes

Median time to extubation

MIDEX: 101 hours (IQR, 65–313 hours) vs. 147 hours (IQR, 81–325 hours) [P=0.01]

PRODEX: 69 hours (IQR, 39–184 hours) vs. 93 hours (IQR, 45–286 hours) [P=0.04]

Median length of stay in ICU

MIDEX: 211 hours (IQR, 115–831 hours) vs. 243 hours (IQR, 140–630 hours) [P=0.27]

PRODEX: 164 hours (IQR, 90–480 hours) vs. 185 hours (IQR, 93–520 hours) [P=0.54]

Ability to communicate pain (visual analogue scale, higher score is better outcome)

MIDEX: 46.3 vs. 24.2 (22.1; 95% CI 17.1–27.1; P<0.001)

PRODEX: 49.3 vs. 35.4 (13.9; 95% CI 8.7–19.1; P<0.001)

Arousability of the patient (VAS)

MIDEX: 58.2 vs. 40.7 (17.5; 95% CI 12.7–22.3; P<0.001)

PRODEX: 59.1 vs. 47.8 (11.2; 95% CI 6.5–16.0; P<0.001)

Cooperation of the patient (VAS)

MIDEX: 44.8 vs. 25.1 (19.7; 95% CI 14.8–24.5; P<0.001)

PRODEX: 47.2 vs. 38.0 (9.2; 95% CI 3.9–14.5; P<0.001)

Subgroup Analysis

Duration of infusion, h

MIDEX: 42 (IRQ 23-72) vs. 43 (IRQ 24-92) [P=0.15]

PRODEX: 42 (IRQ 22-72) vs. 47 (IRQ 25-103) [P=0.001]

RASS score during drug infusion

MIDEX: -0.9 (-1.9 to -0.1) vs. -1.5(-2.5 to 0.5) [P<0.001]

PRODEX: -1.0 (-1.9 to -0.2) vs. -1.7 (-2.5 to -0.7) [P<0.001]

Patients receiving rescue sedation

MIDEX: 4.8% vs. 45.3% [P=0.72]

PRODEX: 72.5% vs. 64.4% [P=0.05]

Spontaneous breathing trial attempted

MIDEX: 58.8% vs. 46.6% [P<0.001]

PRODEX: 58.8% vs. 51.4% [P=0.02]

Adverse Events

Discontinuation due to lack of efficacy

MIDEX: 9% vs. 4% [P=0.02]

PRODEX: 14% vs. 5% [P<0.001]

Hypotension

MIDEX: 20.6% vs. 11.6% [P=0.007]

Bradycardia

MIDEX: 14.2% vs. 5.2% [P<0.001]

First degree AV block

PRODEX: 3.7% vs. 0.8% [P=0.04]

Contraindications to performing sedation stop

Severe cardiovascular instability

PRODEX: 5.8% vs. 3.2% [P=0.02]

Criticisms

• Dexmedetomidine was discontinued due to lack of efficacy more frequently than midazolam or propofol
  • Midazolam and propofol groups achieved deeper mean sedation than dexmedetomidine
  • Dexmedetomidine can be expected to be ineffective in 1 of every 8-10 patients.
• Dexmedetomidine doses were higher and treatment duration was longer than what was recommended
• Internal Validity may be unclear as dexmedetomidine doses were much lower in MIDEX than PRODEX despite targeting the same sedation range
• Internal vality challenging to interpret as mechanical ventilation weaning and extubation criteria not standardized
• Spontaneous breathing trials were inconsistent during sedation vacations
• Dexmedetomidine resulted in shorter times to extubation with no difference in ICU or hospital length of stay
  • Trials may have been under-powered to find difference in length of stay
• Internal validity and External validity challenging with unclear non-inferiority/superiority design of the study.
  • The only outcome analyzed for non-inferiority was “ratio of time at target sedation without rescue medication”.
  • All other outcomes were analyzed for superiority
• Findings can only be applied to patients requiring light-moderate sedation
• Delirium was not routinely measured affecting the external validity.

Funding

Orion Pharma