Perioperative medicine & outcomes
Effectiveness of drug interventions to prevent delirium after surgery for older adults: systematic review and network meta-analysis of randomised controlled trials
An erratum to this article was published (BMJ, 13 Mar 2026 ↗)
Authors' conclusion
Dexmedetomidine is effective in the prevention of postoperative delirium. This finding remains after exclusion of studies at high risk of bias.
Study context
- Population
- Adults ≥60 years having surgery under general or RA
- Participants
- 41,084 adults · 158 trials
- Comparison
- 52 drug interventions for prevention of postoperative delirium
- Outcome
- Postoperative delirium (validated delirium assessment tool)
Methodological detailsBayesian arm based network meta-analysis, Cochrane risk of bias tool version 2, CINeMA
- Bayesian arm based network meta-analysis
- Cochrane risk of bias tool version 2
- CINeMA
Pooled result
As reported in the abstract
Postoperative delirium (trials not at high risk of bias)
Odds ratio (95% CrI)
Dexmedetomidine
0.46(0.36–0.57)
Corticosteroids
0.53(0.31–0.87)
Melatonin receptor agonists
0.54(0.34–0.85)
Parecoxib
0.34(0.16–0.74)
Olanzapine
0.27(0.07–0.94)
Intranasal insulin
0.13(0.04–0.34)
- Overall risk of postoperative delirium
- 14.5% (n=5957)
Further pooled results
Delirium severity (Memorial Delirium Assessment Scale)
| Mean difference (Memorial Delirium Assessment Scale points) (95% CrI) | ||
|---|---|---|
| CorticosteroidsCrI; only intervention reducing severity · CrI | −2.42 (−4.72 to −0.12) | |
| −6−4−202Difference, Memorial Delirium Assessment Scale points (linear scale) |
LOS, mortality, cognition, quality of life
- Most interventions had no effect
Safety
Hypotension, bradycardia, PONV with dexmedetomidine
- Hypotension and bradycardia more common with dexmedetomidine; PONV reduced
Postoperative infection with corticosteroids
- Not increased by corticosteroids
Original publication
Effectiveness of drug interventions to prevent delirium after surgery for older adults: systematic review and network meta-analysis of randomised controlled trials. BMJ. Published 12 February 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →