MENDS: Dexmedetomidine vs Lorazepam for Mechanical Ventilation

Dexmedetomidine more days alive without delirium/coma than benzodiazepines.

Population

103 mechanically ventilated medical/surgical ICU adults expected to need ≥24 hr of sedation.

Intervention

Dexmedetomidine 0.15-1.5 mcg/kg/hr titrated to RASS 0 to -2.

Comparison

Lorazepam infusion 1-10 mg/hr titrated to same RASS goal.

Outcome

Days alive without delirium or coma over 12 days; secondary: ventilator days, mortality.

Double-blind RCT at 2 academic centers. Both arms received concurrent fentanyl PRN. Daily delirium screening with CAM-ICU; sedation with RASS. Open-label rescue with propofol or fentanyl as needed.

• Delirium-/coma-free days: 7.0 dex vs 3.0 lorazepam (P=0.01).
• Days alive on study: 9.5 vs 9.0 (NS).
• ICU mortality numerically lower with dex (17% vs 27%) but underpowered.
• More dexmedetomidine recipients reached RASS goal (80% vs 67%).

Default to non-benzodiazepine sedation (dexmedetomidine or propofol) for mechanically ventilated adults — SCCM PADIS 2018 codified this. Reserve benzodiazepines for refractory agitation, alcohol/benzo withdrawal, status epilepticus. The downstream MENDS-II + SEDCOM trials reinforced these findings; modern ICU practice has moved away from continuous benzo sedation.

• Small sample (103) — adequately powered only for the primary surrogate.
• Single dexmedetomidine dose range; could not address very-deep sedation needs.
• Cost: dex was substantially more expensive at the time (now generic, much less so).
• Did not study post-discharge cognitive outcomes (later trials did).

1. Is your ICU still using midazolam infusions for routine vent sedation? What's the institutional barrier to switching?
2. When is benzodiazepine sedation still appropriate (alcohol withdrawal, refractory status, severe AWS)?
3. How do you handle the bradycardia + hypotension that limit dex titration in some patients?