Esketamine and postoperative delirium: a real signal on soft evidence

Design

systematic review and meta-analysis of 17 randomised controlled trials with GRADE certainty assessment

Population

2,914 surgical patients aged 60 or over; 11 trials in the delirium analysis

Primary outcome

incidence of postoperative delirium

Effect

OR 0.57 (95% CI 0.40 to 0.82), P = 0.002, I squared 45.8%

A systematic review pooled 17 randomised trials and 2,914 patients aged 60 or over having surgery, with 11 trials contributing to the delirium analysis. Esketamine-containing perioperative regimens were associated with lower postoperative delirium, odds ratio 0.57 (95% CI 0.40 to 0.82, P = 0.002, I squared 45.8%).

Secondary findings pointed the same way. Four trials of postoperative cognitive dysfunction or delayed neurocognitive recovery suggested a reduction, odds ratio 0.39 (0.22 to 0.71), though certainty was very low. Nausea and vomiting fell, odds ratio 0.49 (0.32 to 0.75), and psychiatric adverse events did not rise, odds ratio 1.41 (0.67 to 2.96). An exploratory exposure-response analysis found no association between cumulative esketamine dose and delirium risk - which argues against a simple dose-dependent mechanism and is a reason for caution about the whole effect.

What to do with this depends on your starting point. It is not grounds to add esketamine to every older patient's anaesthetic: heterogeneity in what the control arms received was substantial, GRADE certainty was limited, and the absent exposure-response relationship is the kind of inconsistency that precedes a null confirmatory trial. It is grounds to prefer an esketamine-containing regimen where you were already choosing between analgesic adjuncts in a patient at high delirium risk - the nausea benefit and the absence of a psychiatric signal make it a reasonable tiebreaker. Then measure delirium properly with a validated tool, because a unit that adopts this without measuring will never know whether it helped.

• Consider esketamine as the adjunct of choice where you are already choosing one, in high delirium risk
• Do not add it routinely to older patients' anaesthetics on this certainty of evidence
• Screen postoperatively with a validated delirium tool, at least daily for the first three days
• Reassure about psychiatric adverse events - the pooled estimate did not rise
• Expect the nausea and vomiting benefit, which was the most consistent secondary finding

Why it matters

This is the strongest pooled signal yet for a drug intervention against postoperative delirium, in a field where almost nothing pharmacological has worked.

Don't overread it

Most estimates carry low or very low GRADE certainty, and the lack of any dose-response relationship is a real argument against a causal effect.

The statistics, in plain English

An odds ratio of 0.57 is about a 43% lower odds of delirium, and an interval of 0.40 to 0.82 excludes no effect - but I squared of 45.8% means the trials were moderately inconsistent, and the control arms differed in what they actually received. The missing exposure-response relationship matters: if a drug prevents delirium, more of it would usually do more, and the absence of that pattern raises the possibility the association is driven by something other than the esketamine. Very low certainty on the cognitive dysfunction outcome means that figure should not be quoted to patients.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.