ICU sedation, delirium & recovery
Haloperidol in treating delirium, reducing mortality, and preventing delirium occurrence: Bayesian and frequentist meta-analyses
Authors' conclusion
Our study supported the use of haloperidol for delirium treatment in adult ICU patients, but not for delirium prevention.
Study context
- Population
- Adult intensive care unit (ICU) patients (delirium treatment or prevention)7 randomized controlled trials (RCTs) treatment, 5 RCTs prevention
- Comparison
- Haloperidol vs placebo for treatment and prevention of delirium
- Outcome
- Treatment: all-cause mortality and SAEs; prevention: incident delirium, all-cause mortality and SAEs
Methodological detailsBayesian and frequentist analyses, double-blind RCTs only
- Bayesian and frequentist analyses
- Double-blind RCTs only
Pooled result
As reported in the abstract
Delirium treatment (7 RCTs, n = 1767) and prevention (5 RCTs, n = 2509): primary outcomes, Bayesian
- Treatment: probability of CIB (RD < −0.02) in reducing all-cause mortality
- 68%
- Treatment: probability of CIH (RD > 0.02) in causing serious adverse events (SAEs)
- 2%
- Prevention: probability of CIB in all-cause mortality
- 12%
- Prevention: probability of CIB in delirium incidence
- 34%
- Prevention: probability of CIB in SAEs
- 0%
Further pooled results
| Outcome | RD (CI) | |
|---|---|---|
| Rescue benzodiazepine use (delirium treatment)I2 = 0%; frequentist | −0.05 (−0.09 to −0.00) | |
| −0.1−0.0500.05Difference (linear scale) |
- Rescue benzodiazepine use (delirium treatment) — Probability of CIB (RD < −0.02) in reducing rescue benzodiazepine use
- 78%
Efficacy outcomes (delirium prevention)
- Posterior probabilities of CIB across all efficacy outcomes low
- all < 50%
Safety
QTc prolongation (delirium prevention)
- Probability of CIH (risk ratio > 1.1)
- 65%
Other safety outcomes (delirium treatment)
- Probabilities of CIH (RD > 0.02) low
- all < 50%
Original publication
Haloperidol in treating delirium, reducing mortality, and preventing delirium occurrence: Bayesian and frequentist meta-analyses. Crit Care. Published 20 March 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →