Ketamine-based ICU sedation and patient-centered outcomes: A systematic review and meta-analysis

Bottom line

Ketamine-based ICU sedation was associated with significantly shorter ICU length of stay and lower delirium incidence in this meta-analysis, but the delirium finding rests on only 2 of 6 pooled trials and mechanical ventilation duration did not reach statistical significance.

PICO at a glance

• Population: Adult ICU patients enrolled across 6 randomized or prospective controlled trials (903 patients total); five trials enrolled general ICU populations and one enrolled medical, surgical and transplant ICU patients.
• Intervention or exposure: Continuous IV ketamine or esketamine infusion; dosing across the 6 included trials ranged from a 0.5 to 2.5 mg/kg bolus followed by 0.125 to 0.8 mg/kg/h, with Amer 2021 using 1 to 2 mcg/kg/min. Concomitant medications across the trials included propofol, midazolam, dexmedetomidine, remifentanil and fentanyl; the control condition is described only as standard analgosedation.
• Comparator and context: The six included prospective/randomized studies evaluated intravenous ketamine or esketamine as an adjunct to usual ICU analgosedation, most commonly propofol-, benzodiazepine-, or opioid-based protocols. The comparator was usual sedation under protocols that varied by trial; do not state a universal control regimen or target. Table medication lists should be read as concomitant-treatment descriptions, not as a reliable treatment-versus-control inventory for every trial. The pooled analysis found no statistically significant reduction in mechanical-ventilation duration (six studies; MD −0.30 days, 95% CI −1.53 to 0.92; I² 88%), while ICU length of stay was shorter in five studies (MD −0.86 days, 95% CI −1.51 to −0.22; I² 0%). These heterogeneous protocol-level findings do not specify a universal ketamine regimen.
• Outcomes: Mechanical ventilation duration (all 6 studies, 903 patients), ICU length of stay (5 of 6 studies), and delirium incidence (only 2 of 6 studies).
• Study design: PRISMA-guided systematic review and meta-analysis, random-effects models.

Clinical question and design

Ketamine has drawn growing interest as an adjunct ICU sedative because of pharmacologic properties, analgesic effect, relative preservation of respiratory drive and hemodynamics, and an NMDA-receptor mechanism distinct from GABAergic agents like propofol and benzodiazepines, that could theoretically reduce sedative/opioid requirements and downstream complications. Following PRISMA guidelines, the authors conducted a comprehensive literature search through June 2025. Eligible studies were randomized or prospective controlled trials in adult ICU patients comparing continuous IV ketamine or esketamine infusion against standard sedation regimens.

Intervention and outcomes measured

Pooled mean differences (MDs) for continuous outcomes and odds ratios (ORs) for binary outcomes, each with 95% confidence intervals, were calculated using random-effects models. Six studies comprising 903 ICU patients met inclusion criteria. Mechanical ventilation duration (MVD) was reported in all six studies; ICU length of stay (LOS) was reported in five studies; delirium incidence was reported in only two studies, a materially smaller evidence base for that outcome than for the other two. The six trials are Perbet 2018, Amer 2021, Mishra 2024, Mubunda 2024, Li 2025 and Qiao 2025, with per-trial dosing given in the full text.

Results

Mechanical ventilation duration trended shorter with ketamine but did not reach statistical significance (MD, -0.30; 95% CI, -1.53 to 0.92), the interval crossing the null; this outcome had the broadest evidence base of the three (all 6 studies, 903 patients). This was the best-powered outcome in the review, yet it is the one that did not clear significance.

Additional findings

Ketamine-based sedation was associated with a statistically significant reduction in ICU LOS (MD, -0.86; 95% CI, -1.51 to -0.22), pooled from five of the six included studies; the paper states that mechanical ventilation duration and ICU length of stay were measured in calendar days, so the unit is confirmed. Ketamine was also associated with a statistically significant decrease in delirium incidence (OR, 0.55; 95% CI, 0.43-0.72), but this outcome was pooled from only two of the six included studies, a considerably thinner evidence base than the confident-looking OR might suggest on its own.

Uncertainty and limitations

The evidence base is small overall, only 6 studies and 903 patients, and just 2 studies for the delirium outcome, the review's most striking finding. The abstract reports none of the trial identities, dosing regimens or comparator agents, but the full text was obtained on 2026-08-24 and supplies all three, along with risk-of-bias ratings and heterogeneity statistics. The authors' own conclusion explicitly calls for further trials to optimize dosing and confirm long-term benefit.

Literature review and evidence synthesis

This ketamine SR/MA sits within a broader ICU-sedation-strategy literature that has moved away from deep, benzodiazepine-heavy regimens toward lighter, more protocolized approaches. Cochrane reviews on protocol-directed sedation to reduce mechanical ventilation duration (PMID 25562750, updated in PMID 30480753) and on daily sedation interruption (PMID 25005604) established the strategy-level evidence this trial-level ketamine question builds on. A Critical Care Medicine review comparing benzodiazepine versus nonbenzodiazepine-based sedation (PMID 23989093) and a JBI review of dexmedetomidine versus propofol in cardiac surgery patients (PMID 29762314) address the same sedative-choice question from adjacent angles. A Cochrane review of alpha-2 agonists for long-term ICU sedation (PMID 25879090) covers the drug class most directly positioned as ketamine's comparator/adjunct partner in multimodal regimens.

How this fits with the broader evidence

• Current SR/MA (direct evidence): Ketamine-based sedation was associated with significantly shorter ICU LOS and lower delirium, but the delirium finding rests on only 2 of 6 studies, and the best-powered outcome (ventilation duration, all 6 studies) did not reach significance, a hypothesis-supporting signal, not a practice-changing mandate.
• Cochrane sedation-strategy reviews (background evidence, PMID 25562750 and PMID 25005604): These established the strategy-level sedation evidence (protocol-directed sedation and daily sedation interruption) that this trial-level, drug-specific ketamine question builds on.
• Alpha-2 agonist Cochrane review (related evidence, PMID 25879090): Covers alpha-2 agonists, the drug class most directly positioned as ketamine's comparator or adjunct partner in a multimodal, opioid-sparing sedation strategy, giving context for where ketamine might eventually fit if larger trials confirm this signal.

ACPE UAN: 0683-0000-26-036-H01-P

---

Study source: PMID:42177858