Haloperidol for Delirium

Examined in three major trials for the treatment of delirium in ICU patients.

In summary, haloperidol is not shown to have meaningful effects on ICU/hospital LOS, mortality (except for AID-ICU 2022) in the prevention or treatment of hypoactive and hyperactive delirious ICU patients. No trials have closely looked at the symptoms of delirium (distress, agitation, safety of patient care). No significant signal of harm including EPS or arrhythmia from QTc elongation.

In accordance with most recent PADIS (2018) guidelines, routine use of haloperidol does not meaningfully impact LOS. Unclear whether it facilitates patient care and safety.

HOPE-ICU (Lancet 2014)

In critically ill patients does early treatment with haloperidol decrease the time that survivors spend in delirium or coma?

Methods (Summary)

• Double-blind, placebo controlled RCT. Single-centre in general adult ICU in the UK. N=142
• Population: mechanically ventilated patients on PPF/FENT infusions for sedation with target RASS 0 to -1. Could also be treated dwith open-label haloperidol IV, up to 10 mg/day for RASS 2+ or above (all arms)
• Intervention: Haloperidol 2.5 mg q8H scheduled
• Control: placebo
• Outcomes: Number of delirium-free, and coma-free days in 1st 14 days

Results

• No significant difference of days free of coma or delirium at 14 days.
• No significant differences in any mortality, LOS outcomes. No differences in complications including oversedation, QTc > 500 ms, or EPS.
• Trend towards lower sedative/analgesic use, and RASS >1

Takeaway

The use of early regular haloperidol did not prevent patients developing delirium. It did reduce agitation and there was a trend towards decreased use of sedatives and analgesics when haloperidol was given. Other studies will need to tell us if haloperidol is a useful treatment once delirium has developed.

MIND-USA (NEJM 2018)

In delirious adult ICU patients, does the administration of haloperidol or ziprasidone reduce the duration of delirium when compared with placebo?

Population

• N=566 adult ICU patients with acute respiratory failure or shock (NOT clear why others were excluded), delirium (hyperactive or hypoactive; CAM-ICU positive)
• 16 centres in the USA
• exclusion: QTc > 550 ms, Hx NMS, outpatient antipsychotic or lithium treatment, severe baseline cognitive impairment

Intervention and Control

• Interventions: IV haloperidol (2.5 mg if younger than 70, 1.25 mg if older than 70), up to 10 mg per administration and 20 mg per day; OR IV ziprasidone.
• Control: IV placebo
• Schedule: Q12H, dosages were doubled if the patient was still delirious. If the patient was not delirious on two consecutive CAM-ICU assessments then the dosages were halved.

Measures

1. days alive without delirium or coma
2. duration of delirium, time to liberation from ventilation, time to successful ICU discharge, time to hospital discharge, 30-day and 90-day survival
3. safety end-points: TdP, NMS, EPS

Results

• No statistical difference in the adjusted median number of days alive and without coma
• No statistical difference in 30- and 90-day mortality, liberation from ventilation, LOS for ICU and hospital
• Safety: LQTS more common in ziprasidoe than in haloperidol or placebo group.

Takeaways

1. IV haloperidol/zipradisone do not treat delirium with clinically meaningful outcomes in ICU patients. However, the vast majority of patients (88-91%) had hypoactive delirium which is not treated this way anyways
2. IV haloperidol/ziprasidone are safe compared to placebo.
3. More evidence is required for treatment of hyperactive delirium

AID-ICU (NEJM 2022)

In delirious adult ICU patients, does scheduled + PRN IV haloperidol improve mortality and hospital LOS?

https://twitter.com/PulmCrit/status/1585351146482593792

Methods

• N=1000 patients randomized to haloperidol 2.5 mg IB TID + PRN to max 20 mg/d vs placebo
• CAM-ICU/ICDSC positive, 16 centres in Europe. General ICUs.
• Primary outcome: number of days alive and out of the hospital at 90 days

Population

• 64% mechanically ventilated, 15% on RRT
• 55% hypoactive, 45% hyperactive delirium
• median dose haloperidol 8 mg daily

Results

• No significant difference in number of days alive and out of the hospital at 90 days
• Slight decrease in 90-day mortality (36.3 vs 43.3%, RR 0.84 [0.72 to 0.98])
• No significant difference in hospital LOS
• No significant difference in days free of delirium, coma, mechanical ventilation
• No significant increase in adverse events including QTc prolongation, physical restraint use

Takeaways

• Haloperidol is safe, but not clearly effective for treatment of delirium in the ICU (for hyperactive or hypoactive delirium). No effect on days alive and out of hospital at 90 days. Perhaps a small trend towards days free of delirium or coma. Suggest a small mortality benefit.
• trial had WAY more hyperactive delirium than MIND-USA